SimPy Documentation

SimPy DocumentationRelease 2.3.3

Klaus Müller, Tony Vignaux, Ontje Lünsdorf, Stefan Scherfke

Feb 24, 2018

Contents

1 Getting Started 3

2 Manuals 13

3 SimPy Classic Tutorials 235

4 Interfacing to External Packages 321

5 SimPy Classic Tools 335

6 Acknowledgments 361

7 Indices and tables 363

i

ii

SimPy Documentation, Release 2.3.3

SimPy is a process-based discrete-event simulation language based on standard Python. It provides the modeller withcomponents of a simulation model including processes, for active components like customers, messages, and vehicles,and resources, for passive components that form limited capacity congestion points like servers, checkout counters,and tunnels. It also provides monitor variables to aid in gathering statistics. Random variates are provided by thestandard Python random module.

It is based on ideas from Simula and Simscript and provides efficient implementation of co-routines using Python’sgenerators capability. It requires Python 2.7 or later including Python 3.x. It was first released in 2002 under the GNULGPL.

Contents:

Contents 1


2 Contents

CHAPTER 1

Getting Started

1.1 Installation

This file describes the installation of SimPy 2.3.3.

1. Check that you have Python 2.7 or above.

2. You can install SimPy easily via PIP

$ pip install SimPy

If SimPy is already installed, use the -U option for pip to upgrade:

$ pip install -U SimPy

Remember, on Linux/MacOS/Unix you may need root privileges to install SimPy. This also applies to theinstalling SimPy manually, as described below.

3. To manually install a SimPy tarball, or to execute the examples, download and unpack the SimPy archive into afolder (using option “Use folder names” in WinZip, “Re-create folders” in Linux Archive Manager, or similaroption in your unpacker). This will create a SimPy-2.3.3 folder with all source code and documentation.

Open a terminal, cd to the SimPy folder and execute setup.py or pip install .:

$ cd where/you/put/simpy/SimPy-x.y$ python setup.py install$ # or$ pip install .

If you do not have permissions to perform the installation as root, you can install SimPy into a non-standardfolder:

$ cd where/you/put/simpy/SimPy-x.y$ python setup.py install --home <dir>

3

https://pypi.python.org/pypi/pip


4. Run one or more of the programs under docs/examples to see whether Python finds the SimPy module. If youget an error message like ImportError: No module named SimPy, move the SimPy folder into a directory whichyou know to be on the Python module search path (like /Lib/site-packages).

5. The tutorial and manuals are in the docs/html folder. Many users have commented that the Bank tutorials arevaluable in getting users started on building their own simple models. Even a few lines of Python and SimPycan model significant real systems.

For more help, contact the SimPy-Users mailing list. SimPy users are pretty helpful.

Enjoy simulation programming in SimPy!

1.2 Contents of This SimPy Distribution

SimPy 2.3.3 contains the following files:

• SimPy - Python code directory for the SimPy 2.3.3 package

– Lister.py, a prettyprinter for class instances

– Simulation.py, code for SimPy simulation

– SimulationTrace.py, code for simulation with tracing

– SimulationStep.py, code for executing simulations event-by-event

– SimulationRT.py, code for synchronizing simulation time with wallclock time

– SimulationGUIDebug.py, code for debugging/event stepping of models with a GUI

– SimGUI.py, code for generating a Tk-based GUI for SimPy simulations

– SimPlot.py, code for generating Tk-based plots (screen and Postscript)

– __init__.py, initialisation of SimPy package

• Tests - a directory containing tests for simpy

• docs - a directory containing the complete, browseable (HTML) documentation of SimPy. It includes tu-torials and descriptions of accessing external packages from SimPy. Click on index.html!

• docs/examples - some SimPy models (in traditional and Object Oriented API)

– Bankmodels - a sub-directory with the models of the Bank tutorials (in traditional and Object OrientedAPI)

• LICENSE.txt - GNU Lesser General Public Licence text

1.3 Changes from Release 2.3.3

This section addresses the difference between previous SimPy version 2.1.0 and 2.3.3 in terms of changes and addi-tions.

1.3.1 Changes from 2.1.0 to 2.2.b1

• The Unit tests have been rewritten

• The directory sturcture of the release has been simplified

• The documentation has had some minor changes

4 Chapter 1. Getting Started

mailto:[email protected]


1.4 COMPATIBILITY: SimPy

SimPy has been used successfully with many packages and modules, such as Tk/Tkinter for GUIs and VPython andmatplotlib for graphical output.

The design of SimPy is such that no incompatibilities with Python 2.7 through 3.6 modules or Python 2.7 through3.6-accessible packages are expected.

SimPy 2.3.3 has been tested with Python 2.7, 3.4, 3.5, 3.6. On Linux and Windows 10.

Should SimPy users discover any incompatibilities, the authors would be grateful for a report. Just send a messagewith the problem and its context to: [email protected].

1.5 SimPy History

SimPy is based on ideas from Simula and Simscript but uses standard Python. It combines two previous packages,SiPy, in Simula-Style (Klaus Muller) and SimPy, in Simscript style (Tony Vignaux and Chang Chui)

SimPy is based on efficient implementation of co-routines using Python’s generators capability.

The package has been hosted on Sourceforge.net since 15 September 2002. Sourceforge.net’s service has always beenoutstanding. It is essential to the SimPy project! Thanks, all you people at SourceForge!

1.5.1 December 2011: Release 2.3

• Support for Python 3.x has been added

• Examples and tutorials modified to run on Python 2.6 and up including Python 3.

• Examples can now be executed via py.test so we can make sure they do run.

• The documentation has had some reorganisation. The index has had work done on it. The Simple manual hasbeen pulled out and is setup as a separate manual.

1.5.2 September 2011: Release 2.2b1

• The Unit tests have been rewritten.

• The directory sturcture of the release has been simplified


1.5.3 May 2010: Version 2.1.0

A major release of SimPy, with a new code base, a (small) number of additions to the API, and added documentation.

Additions

• A function step has been added to the API. When called, it executes the next scheduled event. (step is actuallya method of Simulation.)

• Another new function is peek. It returns the time of the next event. By using peek and step together, one caneasily write e.g. an interactive program to step through a simulation event by event.

1.4. COMPATIBILITY: SimPy 5



• A simple interactive debugger stepping.py has been added. It allows stepping through a simulation, withoptions to skip to a certain time, skip to the next event of a given process, or viewing the event list.

• Versions of the Bank tutorials (documents and programs) using the advanced object-oriented API have beenadded.

• A new document describes tools for gaining insight into and debugging SimPy models.

Changes

• Major re-structuring of SimPy code, resulting in much less SimPy code – great for the maintainers.

• Checks have been added which test whether entities belong to the same Simulation instance.

• The Monitor and Tally methods timeAverage and timeVariance now calculate only with the observed time-series.No value is assumed for the period prior to the first observation.

• Changed class Lister so that circular references between objects no longer lead to stack overflow and crash.

Repairs

• Functions allEventNotices and allEventTimes are working again.

• Error messages for methods in SimPy.Lib work again.

1.5.4 April 2009: Release 2.0.1

A bug-fix release of SimPy 2.0

1.5.5 October 2008: Version 2.0

This is a major release with changes to the SimPy application programming interface (API) and the formatting of thedocumentation.

API changes

In addition to its existing API, SimPy now also has an object oriented API. The additional API

• allows running SimPy in parallel on multiple processors or multi-core CPUs,

• supports better structuring of SimPy programs,

• allows subclassing of class Simulation and thus provides users with the capability of creating new simulationmodes/libraries like SimulationTrace, and

• reduces the total amount of SimPy code, thereby making it easier to maintain.

Note that the OO API is in addition to the old API. SimPy 2.0 is fully backward compatible.

Documentation format changes

SimPy’s documentation has been restructured and processed by the Sphinx documentation generation tool. This hasgenerated one coherent, well structured document which can be easily browsed. A seach capability is included.



1.5.6 March 2008: Version 1.9.1

This is a bug-fix release which cures the following bugs:

• Excessive production of circular garbage, due to a circular reference between Process instances and event no-tices. This led to large memory requirements.

• Runtime error for preempts of proceeses holding multiple Resource objects.

It also adds a Short Manual, describing only the basic facilities of SimPy.

1.5.7 December 2007: Version 1.9

This is a major release with added functionality/new user API calls and bug fixes.

Major changes

• The event list handling has been changed to improve the runtime performance of large SimPy models (modelswith thousands of processes). The use of dictionaries for timestamps has been stopped. Thanks are due to Prof.Norm Matloff and a team of his students who did a study on improving SimPy performance. This was oneof their recommendations. Thanks, Norm and guys! Furthermore, in version 1.9 the ‘heapq’ sorting packagereplaces ‘bisect’. Finally, cancelling events no longer removes them, but rather marks them. When their eventtime comes, they are ignored. This was Tony Vignaux’ idea!

• The Manual has been edited and given an easier-to-read layout.

• The Bank2 tutorial has been extended by models which use more advanced SimPy commands/constructs.

Bug fixes

• The tracing of ‘activate’ statements has been enabled.

Additions

• A method returning the time-weighted variance of observations has been added to classes Monitor and Tally.

• A shortcut activation method called “start” has been added to class Process.

1.5.8 January 2007: Version 1.8

Major Changes

• SimPy 1.8 and future releases will not run under the obsolete Python 2.2 version. They require Python 2.3 orlater.

• The Manual has been thoroughly edited, restructured and rewritten. It is now also provided in PDF format.

• The Cheatsheet has been totally rewritten in a tabular format. It is provided in both XLS (MS Excel spreadsheet)and PDF format.

• The version of SimPy.Simulation(RT/Trace/Step) is now accessible by the variable ‘version’.

• The __str__ method of Histogram was changed to return a table format.

1.5. SimPy History 7


Bug fixes

• Repaired a bug in yield waituntil runtime code.

• Introduced check for capacity parameter of a Level or a Store being a number > 0.

• Added code so that self.eventsFired gets set correctly after an event fires in a compound yield get/put with awaitevent clause (reneging case).

• Repaired a bug in prettyprinting of Store objects.

Additions

• New compound yield statements support time-out or event-based reneging in get and put operations on Storeand Level instances.

• yield get on a Store instance can now have a filter function.

• All Monitor and Tally instances are automatically registered in list allMonitors and allTallies, respectively.

• The new function startCollection allows activation of Monitors and Tallies at a specified time.

• A printHistogram method was added to Tally and Monitor which generates a table-form histogram.

• In SimPy.SimulationRT: A function for allowing changing the ratio wall clock time to simulation time has beenadded.

1.5.9 June 2006: Version 1.7.1

This is a maintenance release. The API has not been changed/added to.

• Repair of a bug in the _get methods of Store and Level which could lead to synchronization problems (blockingof producer processes, despite space being available in the buffer).

• Repair of Level __init__ method to allow initialBuffered to be of either float or int type.

• Addition of type test for Level get parameter ‘nrToGet’ to limit it to positive int or float.

• To improve pretty-printed output of ‘Level’ objects, changed attribute ‘_nrBuffered’ to ‘nrBuffered’ (synonymfor ‘amount’ property).

• To improve pretty-printed output of ‘Store’ objects, added attribute ‘buffered’ (which refers to ‘_theBuffer’attribute).

1.5.10 February 2006: Version 1.7

This is a major release.

• Addition of an abstract class Buffer, with two sub-classes Store and Level Buffers are used for modelling inter-process synchronization in producer/ consumer and multi-process cooperation scenarios.

• Addition of two new yield statements:

– yield put for putting items into a buffer, and

– yield get for getting items from a buffer.

• The Manual has undergone a major re-write/edit.



• All scripts have been restructured for compatibility with IronPython 1 beta2. This was doen by moving allimport statements to the beginning of the scripts. After the removal of the first (shebang) line, all scripts (withthe exception of plotting and GUI scripts) can run successfully under this new Python implementation.

1.5.11 September 2005: Version 1.6.1

This is a minor release.

• Addition of Tally data collection class as alternative to Monitor. It is intended for collecting very large data setsmore efficiently in storage space and time than Monitor.

• Change of Resource to work with Tally (new Resource API is backwards-compatible with 1.6).

• Addition of function setHistogram to class Monitor for initializing histograms.

• New function allEventNotices() for debugging/teaching purposes. It returns a prettyprinted string with eventtimes and names of process instances.

• Addition of function allEventTimes (returns event times of all scheduled events).

1.5.12 15 June 2005: Version 1.6

• Addition of two compound yield statement forms to support the modelling of processes reneging from resourcequeues.

• Addition of two test/demo files showing the use of the new reneging statements.

• Addition of test for prior simulation initialization in method activate().

• Repair of bug in monitoring thw waitQ of a resource when preemption occurs.

• Major restructuring/editing to Manual and Cheatsheet.

1.5.13 1 February 2005: Version 1.5.1

• MAJOR LICENSE CHANGE:

Starting with this version 1.5.1, SimPy is being release under the GNU Lesser General Public License(LGPL), instead of the GNU GPL. This change has been made to encourage commercial firms to useSimPy in for-profit work.

• Minor re-release

• No additional/changed functionality

• Includes unit test file’MonitorTest.py’ which had been accidentally deleted from 1.5

• Provides updated version of ‘Bank.html’ tutorial.

• Provides an additional tutorial (‘Bank2.html’) which shows how to use the new synchronization constructsintroduced in SimPy 1.5.

• More logical, cleaner version numbering in files.



1.5.14 1 December 2004: Version 1.5

• No new functionality/API changes relative to 1.5 alpha

• Repaired bug related to waiting/queuing for multiple events

• SimulationRT: Improved synchronization with wallclock time on Unix/Linux

1.5.15 25 September 2004: Version 1.5alpha

• New functionality/API additions

– SimEvents and signalling synchronization constructs, with ‘yield waitevent’ and ‘yield queueevent’ com-mands.

– A general “wait until” synchronization construct, with the ‘yield waituntil’ command.

• No changes to 1.4.x API, i.e., existing code will work as before.

1.5.16 19 May 2004: Version 1.4.2

• Sub-release to repair two bugs:

– The unittest for monitored Resource queues does not fail anymore.

– SimulationTrace now works correctly with “yield hold,self” form.

• No functional or API changes

1.5.17 29 February 2004: Version 1.4.1

• Sub-release to repair two bugs:

– The (optional) monitoring of the activeQ in Resource now works correctly.

– The “cellphone.py” example is now implemented correctly.


1.5.18 1 February 2004: Version 1.4 published on SourceForge

1.5.19 22 December 2003: Version 1.4 alpha

• New functionality/API changes

– All classes in the SimPy API are now new style classes, i.e., they inherit from object eitherdirectly or indirectly.

– Module Monitor.py has been merged into module Simulation.py and all SimulationXXX.py mod-ules. Import of Simulation or any SimulationXXX module now also imports Monitor.

– Some Monitor methods/attributes have changed. See Manual!

– Monitor now inherits from list.

– A class Histogram has been added to Simulation.py and all SimulationXXX.py modules.

– A module SimulationRT has been added which allows synchronization between simulated andwallclock time.



– A moduleSimulationStep which allows the execution of a simulation model event-by-event, withthe facility to execute application code after each event.

– A Tk/Tkinter-based module SimGUI has been added which provides a SimPy GUI framework.

– A Tk/Tkinter-based module SimPlot has been added which provides for plot output from SimPyprograms.

1.5.20 22 June 2003: Version 1.3


• Reduction of sourcecode linelength in Simulation.py to <= 80 characters

1.5.21 9 June 2003: Version 1.3 alpha

• Significantly improved performance

• Significant increase in number of quasi-parallel processes SimPy can handle

• New functionality/API changes:

– Addition of SimulationTrace, an event trace utility

– Addition of Lister, a prettyprinter for instance attributes

– No API changes

• Internal changes:

– Implementation of a proposal by Simon Frost: storing the keys of the event set dictionary in a binary searchtree using bisect. Thank you, Simon! SimPy 1.3 is dedicated to you!

• Update of Manual to address tracing.

• Update of Interfacing doc to address output visualization using Scientific Python gplt package.

1.5.22 29 April 2003: Version 1.2

• No changes in API.

• Internal changes:

– Defined “True” and “False” in Simulation.py to support Python 2.2.

1.5.23 22 October 2002:

• Re-release of 0.5 Beta on SourceForge.net to replace corrupted file __init__.py.

• No code changes whatever!

1.5.24 18 October 2002:

• Version 0.5 Beta-release, intended to get testing by application developers and system integrators in preparationof first full (production) release. Released on SourceForge.net on 20 October 2002.

• More models



• Documentation enhanced by a manual, a tutorial (“The Bank”) and installation instructions.

• Major changes to the API:

– Introduced ‘simulate(until=0)’ instead of ‘scheduler(till=0)’. Left ‘scheduler()’ in for backward compati-bility, but marked as deprecated.

– Added attribute “name” to class Process. Process constructor is now:

def __init__(self,name="a_process")

Backward compatible if keyword parameters used.

– Changed Resource constructor to:

def __init__(self,capacity=1,name="a_resource",unitName="units")

Backward compatible if keyword parameters used.

1.5.25 27 September 2002:

• Version 0.2 Alpha-release, intended to attract feedback from users

• Extended list of models

• Upodated documentation

1.5.26 17 September 2002

• Version 0.1.2 published on SourceForge; fully working, pre-alpha code

• Implements simulation, shared resources with queuing (FIFO), and monitors for data gathering/analysis.

• Contains basic documentation (cheatsheet) and simulation models for test and demonstration.

1.6 SimPy Resources

SimPy can be downloaded from the “SimPy web-site”: https://github.com/SimPyClassic/SimPyClassic

Simulation model developers are encouraged to share their SimPy modeling techniques with the SimPy community.

Software developers are encouraged to interface SimPy with other Python-accessible packages, such as GUI, data baseor mapping and to share these new capabilities with the community under the GNU GPL.

Feature requests for future SimPy versions should be sent to “Klaus G. Muller”, kgmuller at users.sourceforge.net, or“Tony Vignaux”, vignaux at users.sourceforge.net.


https://github.com/SimPyClassic/SimPyClassic

CHAPTER 2

Manuals

2.1 SimPy Classic Manual

Authors

• Tony Vignaux <[email protected]>

• Klaus Muller <[email protected]>

• Bob Helmbold

Release 2.3.3

Web-site https://github.com/SimPyClassic/SimPyClassic

Python-Version 2.7 and later

Date Feb 24, 2018

Contents

• SimPy Classic Manual

– Introduction

– Simulation with SimPy

– Processes

– Resources

– Levels

– Stores

– Random Number Generation

– Recording Simulation Results

13





– Other Links

– Acknowledgments

– Appendices

– Glossary

This document describes SimPy Classic version 2.3.3. Changes from the previous version are listed in Appendix A0.

Note: This document does not describe the object oriented (OO) API which has been added to SimPy with version2.0. SimPy 2.0 is fully backward compatible with previous versions. The procedural API and the OO API co-existhappily in SimPy 2.x.

2.1.1 Introduction

SimPy is a Python-based discrete-event simulation system that models active components such as messages, cus-tomers, trucks, planes by parallel processes. It provides a number of tools for the simulation programmer includingProcesses to model active entities, three kinds of resource facilities (Resources, Levels, and Stores) and ways of record-ing results by using Monitors and Tallys.

The basic active elements of a SimPy model are process objects (i.e., objects of a Process class – see Processes). Asa general practice and for brevity we will often refer to both process objects and their classes as “processes.” Thus,“process” may refer to a Process class or to a process object, depending on context. To avoid ambiguity or for addedemphasis we often explicitly state whether a class or an object is intended. In addition we will use “entity” to referto process objects as this is frequently used in the simulation literature. Here, though, we restrict it to process objectsand it will not be used for any other elements in the simulation.

During the simulation, Process objects may be delayed for fixed or random times, queued at resource facilities, andmay be interrupted by or interact in other ways with other processes and components. For example, Automobiles in amodel of a gas station may have to queue while waiting for a pump to become available . Once obtaining a pump ittakes some time to fill before releasing the pump.

A SimPy script contains the declaration of one or more Process classes and the creation of process objects (entities)from them. Each process object executes its Process Execution Method (referred to later as a PEM), a method thatdetermines its actions. Each PEM runs in parallel with (and may interact with) the PEMs of other process objects.

There are three types of resource facilities (Resources, Levels, and Stores). Each type models a congestion point whereprocess objects may have to queue while waiting to acquire or, in some cases to deposit, a resource.

Resources have several resource units, each of which may be used by process objects. Extending the example above,the gas station might be modelled as a resource with its pumps as resource units. On receiving a request for a pumpfrom a car, the gas station resource automatically queues waiting cars until one becomes available. The pump resourceunit is held by the car until it is released for possible use by another car.

Levels model the supply and consumption of a homogeneous undifferentiated “material.” The Level at any time holdsan amount of material that is fully described by a scalar (real or integer). This can be increased or decreased by processobjects. For example, a gas (petrol) station stores gas in large storage tanks. The tanks can be increased by Tankerdeliveries and reduced by cars refuelling. A car need not return the gas to the Level in contrast to the requirement forResource units.

Stores model the production and consumption of individual items. A store hold a list of items. Process objects caninsert or remove items from the list. For example, surgical procedures (treated as process objects) require specific listsof personnel and equipment that may be treated as the items in a Store facility such as a clinic or hospital. The itemsheld in a Store can be of any Python type. In particular they can be process objects, and this may be exploited tofacilitate modelling Master/Slave relationships.

14 Chapter 2. Manuals


In addition to the number of free units or quantities, resource facilities all hold queues of waiting process objects whichare operated automatically by SimPy. They also operate a reneging mechanism so that a process object can abandonthe wait.

Monitors and Tallys are used to compile statistics as a function of time on variables such as waiting times and queuelengths. These statistics consist of simple averages and variances, time-weighted averages, or histograms. They can begathered on the queues associated with Resources, Levels and Stores. For example we may collect data on the averagenumber of cars waiting at a gas station and the distribution of their waiting times. Tallys update the current statisticsas the simulation progresses, but cannot preserve complete time-series records. Monitors can preserve complete time-series records that may later be used for more advanced post-simulation analyses.

Before attempting to use SimPy, you should be able to write Python code. In particular, you should be able to defineand use classes and their objects. Python is free and usable on most platforms. We do not expound it here. You canfind out more about it and download it from the Python web-site (https://www.python.org). SimPy requires Python2.3 or later.

[Return to Top ]

2.1.2 Simulation with SimPy

To use the SimPy simulation system you must import its Simulation module (or one of the alternatives):

from SimPy.Simulation import *

All discrete-event simulation programs automatically maintain the current simulation time in a software clock. Thiscannot be changed by the user directly. In SimPy the current clock value is returned by the now() function.

At the start of the simulation the software clock is set to 0.0. While the simulation program runs, simulation time stepsforward from one event to the next. An event occurs whenever the state of the simulated system changes. For example,an event might be the arrival or departure of a car from the gas station.

The following statement initializes global simulation variables and sets the software clock to zero. It must appear inthe script before any SimPy process objects are activated.

initialize( )

This is followed by SimPy statements creating and activating process objects. Activation of process objects addsevents to the simulation schedule. Execution of the simulation itself starts with the following statement:

simulate(until=endtime)

The simulation starts, and SimPy seeks and executes the first scheduled event. Having executed that event, the simu-lation seeks and executes the next event, and so on.

Typically a simulation is terminated when endtime is reached but it can be stopped at any time by the command:

stopSimulation( )

now( ) will then equal the time when this was called. The simulation will also stop if there are no more events toexecute (so now() equals the time the last scheduled event occurred)

After the simulation has stopped, further statements can be executed. now() will retain the time of stopping and dataheld in Monitors will be available for display or further analysis.

The following fragment shows only the main block in a simulation program. (Complete, runnable examples are shownin Example 1 and Example 2). Here Message is a (previously defined) Process class and m is defined as an object ofthat class, that is, a particular message. Activating m has the effect of scheduling at least one event by starting the PEMof m (here called go). The simulate(until=1000.0) statement starts the simulation itself, which immediatelyjumps to the first scheduled event. It will continue until it runs out of events to execute or the simulation time reaches1000.0. When the simulation stops the (previously written) Report function is called to display the results:

2.1. SimPy Classic Manual 15

https://www.python.org



1 initialize()2 m = Message()3 activate(m, m.go(), at=0.0)4 simulate(until=1000.0)5

6 Report() # report results when the simulation finishes

The object-oriented interface

An object-oriented API interface was added in SimPy 2.0. It is described more fully in SimPyOO_API. It definesa class of Simulation objects and makes running multiple simulations cleaner and easier. It is compatible with theprocedural version described in this Manual. Using the object-oriented API, the program fragment listed at the end ofthe previous subsection would look like this:

1 s = Simulation()2 s.initialize()3 m = Message(sim=s)4 s.activate(m, m.go(), at=0.0)5 s.simulate(until=1000.0)6

7 Report() # report results when the simulation finishes

Further examples of the OO style exist in the SimPyModels directory and the Bank Tutorial.

Alternative SimPy simulation libraries

In addition to SimPy.Simulation, SimPy provides four alternative simulation libraries which have the basic SimPy.Simulation capabilities, plus additional facilities:

• SimPy.SimulationTrace for program tracing: With from SimPy.SimulationTrace import *, anySimPy program automatically generates detailed event-by-event tracing output. This makes the library idealfor program development/testing and for teaching SimPy.

• SimPy.SimulationRT for real time synchronization: from SimPy.SimulationRT import * facilitatessynchronizing simulation time and real (wall-clock) time. This capability can be used to implement, e.g., inter-active game applications or to demonstrate a model’s execution in real time.

• SimPy.SimulationStep for event-stepping through a simulation: The import from SimPy.SimulationStep import * provides an API for stepping through a simulation event by event.This can assist with debugging models, interacting with them on an event-by-event basis, getting event-by-eventoutput from a model (e.g. for plotting purposes), etc.

• SimPy.SimulationGUIDebug for event-stepping through a simulation with a GUI: from SimPy.SimulationGUIDebug import * provides an API for stepping through a simulation event-by-event, witha GUI for user control. The event list, Process and Resource objects are shown in windows. This is useful fordebugging models and for teaching discrete event simulation with SimPy.

[Return to Top ]

2.1.3 Processes

The active objects for discrete-event simulation in SimPy are process objects – instances of some class that inheritsfrom SimPy’s Process class.



For example, if we are simulating a computing network we might model each message as an object of the classMessage. When message objects arrive at the computing network they make transitions between nodes, wait forservice at each one, are served for some time, and eventually leave the system. The Message class specifies allthe actions of each message in its Process Execution Method (PEM). Individual message objects are created as thesimulation runs, and their evolutions are directed by the Message class’s PEM.

Defining a process

Each Process class inherits from SimPy’s Process class. For example the header of the definition of a new MessageProcess class would be:

class Message(Process):

At least one Process Execution Method (PEM) must be defined in each Process class1. A PEM may have argumentsin addition to the required self argument that all methods must have. Naturally, other methods and, in particular, an__init__ method, may be defined.

• A Process Execution Method (PEM) defines the actions that are performed by its process objects.Each PEM must contain at least one of the yield statements, described later. This makes it a Python generatorfunction so that it has resumable execution – it can be restarted again after the yield statement without losing itscurrent state. A PEM may have any name of your choice. For example it may be called execute( ) or run().

“The yield statements are simulation commands which affect an ongoing life-cycle of Process objects. Thesestatements control the execution and synchronization of multiple processes. They can delay a process, put itto sleep, request a shared resource or provide a resource. They can add new events on the simulation eventschedule, cancel existing ones, or cause processes to wait for a state change.”

For example, here is a the Process Execution Method, go(self), for the Message class. Upon activationit prints out the current time, the message object’s identification number and the word “Starting”. After asimulated delay of 100.0 time units (in the yield hold, ... statement) it announces that this messageobject has “Arrived”:

def go(self):print(now(), self.i, 'Starting')yield hold, self, 100.0print(now(), self.i, 'Arrived')

A process object’s PEM starts execution when the object is activated, provided the simulate(until= ...)statement has been executed.

• __init__(self, . . . ), where . . . indicates method arguments. This method initializes the process object, settingvalues for some or all of its attributes. As for any sub-class in Python, the first line of this method must call theProcess class’s __init__( ) method in the form:

Process.__init__(self)

You can then use additional commands to initialize attributes of the Process class’s objects. You can also overridethe standard name attribute of the object.

The __init__( ) method is always called whenever you create a new process object. If you do not wish toprovide for any attributes other than a name, the __init__ method may be dispensed with. An example of an__init__( ) method is shown in the example below.

1 The variable version, imported from SimPy.Simulation, contains the revision number and date of the current version.



Creating a process object

An entity (process object) is created in the usual Python manner by calling the class. Process classes have a singleargument, name which can be specified if no __init__ method is defined. It defaults to 'a_process'. It can beover-ridden if an __init__ method is defined.

For example to create a new Message object with a name Message23:

m = Message(name="Message23")

Note: When working through this and all other SimPy manuals, the reader is encouraged to type in, run andexperiment with all examples as she goes. No better way of learning exists than doing! A suggestion: if youwant to see how a SimPy model is being executed, trace it by replacing from SimPy.Simulation import * with fromSimPy.SimulationTrace import *. Any Python environment is suitable – an interactive Python session, IDLE, IPython,Scite . . .

Example 1: This is is a complete, runnable, SimPy script. We declare a Message class and define an__init__( ) method and a PEM called go( ). The __init__( ) method provide an instancevariables of an identification number and message length. We do not actually use the len attribute in thisexample.

Two messages, p1 and p2 are created. p1 and p2 are activated to start at simulation times 0.0 and6.0, respectively. Nothing happens until the simulate(until=200) statement. When both messageshave finished (at time 6.0+100.0=106.0) there will be no more events so the simulation will stop at thattime:

from SimPy.Simulation import Process, activate, initialize, hold, now,→˓simulate

class Message(Process):"""A simple Process"""

def __init__(self, i, len):Process.__init__(self, name='Message' + str(i))self.i = iself.len = len

def go(self):print('%s %s %s' % (now(), self.i, 'Starting'))yield hold, self, 100.0print('%s %s %s' % (now(), self.i, 'Arrived'))

initialize()p1 = Message(1, 203) # new messageactivate(p1, p1.go()) # activate itp2 = Message(2, 33)activate(p2, p2.go(), at=6.0)simulate(until=200)print('Current time is %s' % now()) # will print 106.0

Running this program gives the following output:

0 1 Starting6.0 2 Starting



100.0 1 Arrived106.0 2 ArrivedCurrent time is 106.0

Elapsing time in a Process

A PEM uses the yield hold command to temporarily delay a process object’s operations.

yield hold

yield hold, self,t

Causes the process object to delay t time units2. After the delay, it continues with the next statement in itsPEM. During the hold the object’s operations are suspended.

Example 2: In this example the Process Execution Method, buy, has an extra argument, budget:

from SimPy.Simulation import Process, activate, hold, initialize, simulate

class Customer(Process):def buy(self, budget=0):

print('Here I am at the shops %s' % self.name)t = 5.0for i in range(4):

yield hold, self, t# executed 4 times at intervals of t time unitsprint('I just bought something %s' % self.name)budget -= 10.00

print('All I have left is %s I am going home %s' % (budget, self.→˓name))

initialize()

# create a customer named "Evelyn",C = Customer(name='Evelyn')

# and activate her with a budget of 100activate(C, C.buy(budget=100), at=10.0)

simulate(until=100.0)

Starting and stopping SimPy Process Objects

A process object is “passive” when first created, i.e., it has no scheduled events. It must be activated to start its ProcessExecution Method. To activate an instance of a Process class you can use either the activate function or the startmethod of the Process. (see the Glossary for an explanation of the modified Backus-Naur Form (BNF) notation used).

2 More than one can be defined but only one can be executed by any process object.



activate

• activate(p, p.pemname([args])[,{at=now()|delay=0}][,prior=False])

activates process object p, provides its Process Execution Method p.pemname( ) with arguments args and pos-sibly assigns values to the other optional parameters. The default is to activate at the current time (at=now()) with no delay (delay=0.0) and prior set to False. You may assign other values to at, delay, andprior.

Example: to activate a process object, cust with name cust001 at time 10.0 using a PEM called lifetime:

activate(cust, cust.lifetime(name='cust001'), at=10.0)

However, delay overrides at, in the sense that when a delay=period clause is included, then activationoccurs at now( ) or now( )+period (whichever is larger), irrespective of what value of t is assigned in theat=t clause. This is true even when the value of period in the delay clause is zero, or even negative. So it isbetter and clearer to choose one (or neither) of at=t and delay=period, but not both.

If you set prior=True, then process object p will be activated before any others that happen to be scheduledfor activation at the same time. So, if several process objects are scheduled for activation at the same time andall have prior=True, then the last one scheduled will actually be the first to be activated, the next-to-last ofthose scheduled, the second to be activated, and so forth.

Retroactive activations that attempt to activate a process object before the current simulation time terminate thesimulation with an error report.

start

An alternative to activate() function is the start method. There are a number of ways of using it:

• p.start(p.pemname([args])[,{at=now()|delay=0}][,prior=False])

is an alternative to the activate statement. p is a Process object. The generator function, pemname, can haveany identifier (such as run, life-cycle, etc). It can have parameters.

For example, to activate the process object cust using the PEM with identifier, lifetime at time 10.0 wewould use:

cust.start(cust.lifetime(name='cust001'), at=10.0)

• p.start([p.ACTIONS()] [,{at=now()|delay=0}][,prior=False])

if p is a Process object and the generator function is given the standard identifier, ACTIONS. ACTIONS, isrecognized as a Process Execution Method. It may not have parameters. The call p.ACTIONS() is optional.

For example, to activate the process object cust with the standard PEM identifier ACTIONS at time 10.0, thefollowing are equivalent (and the second version is more convenient):

cust.start(cust.ACTIONS(), at=10.0)cust.start(at=10.0)

• An anonymous instance of Process class PR can be created and activated in one command using start withthe standard PEM identifier, ACTIONS.

PR.([args]).start( [,{at=now()|delay=0}][,prior=False])

Here, PR is the identifier for the Process class and not for a Process object as was p, in the statements above.The generator method ACTIONS may not have parameters.



For example, if Customer is a SimPy Process class we can create and activate an anonymous instance at time10.0:

Customer(name='cust001').start(at=10.0)

You can use the passivate, reactivate, or cancel commands to control Process objects.

passivate

• yield passivate, self

suspends the process object itself. It becomes “passive”. To get it going again another process mustreactivate it.

reactivate

• reactivate(p[,{at=now()|delay=0}][,prior=False])

reactivates a passive process object, p. It becomes “active”. The optional parameters work as for activate.A process object cannot reactivate itself. To temporarily suspend itself it must use yield hold,self,tinstead.

cancel

• self.cancel(p)

deletes all scheduled future events for process object p. A process cannot cancel itself. If that is required, useyield passivate,self instead. Only “active” process objects can be canceled.

A process object is “terminated” after all statements in its process execution method have been completed. If theobject is still referenced by a variable, it becomes just a data container. This can be useful for extracting information.Otherwise, it is automatically destroyed.

Even activated process objects will not start operating until the simulate(until=endtime) statement is executed.This starts the simulation going and it will continue until time endtime (unless it runs out of events to execute or thecommand stopSimulation( ) is executed).

Example 3 This simulates a firework with a time fuse. We have put in a few extra yield hold commands for addedsuspense.

from SimPy.Simulation import Process, activate, initialize, hold, now, simulate

class Firework(Process):

def execute(self):print('%s firework launched' % now())yield hold, self, 10.0 # wait 10.0 time unitsfor i in range(10):

yield hold, self, 1.0print('%s tick' % now())

yield hold, self, 10.0 # wait another 10.0 time unitsprint('%s Boom!!' % now())



initialize()f = Firework() # create a Firework object, and# activate it (with some default parameters)activate(f, f.execute(), at=0.0)simulate(until=100)

Here is the output. No formatting was attempted so it looks a bit ragged:

0 firework launched11.0 tick12.0 tick13.0 tick14.0 tick15.0 tick16.0 tick17.0 tick18.0 tick19.0 tick20.0 tick30.0 Boom!!

A source fragment

One useful program pattern is the source. This is a process object with a Process Execution Method (PEM) thatsequentially generates and activates other process objects – it is a source of other process objects. Random arrivalscan be modelled using random intervals between activations.

Example 4: A source. Here a source creates and activates a series of customers who arrive at regular intervals of10.0 units of time. This continues until the simulation time exceeds the specified finishTime of 33.0. (Of course,to model customers with random inter-arrival times the yield hold statement would use a random variate, such asexpovariate( ), instead of the constant 10.0 inter-arrival time used here.) The following example assumes thatthe Customer class has previously been defined with a PEM called run that does not require any arguments:

1 class Source(Process):2

3 def execute(self, finish):4 while now() < finish:5 c = Customer() # create a new customer object, and6 # activate it (using default parameters)7 activate(c, c.run())8 print('%s %s) % (now(), 'customer')9 yield hold, self, 10.0

10

11 initialize()12 g = Source() # create the Source object, g,13 # and activate it14 activate(g, g.execute(finish=33.0), at=0.0)15 simulate(until=100)



Asynchronous interruptions

An active process object can be interrupted by another but cannot interrupt itself.

interrupt

• self.interrupt(victim)

The interrupter process object uses its interrupt method to interrupt the victim process object. The interruptis just a signal. After this statement, the interrupter process object continues its PEM.

For the interrupt to have an immediate effect, the victim process object must be active – that is it must have anevent scheduled for it (that is, it is “executing” a yield hold ). If the victim is not active (that is, it is eitherpassive or terminated) the interrupt has no effect. For example, process objects queuing for resource facilitiescannot be interrupted because they are passive during their queuing phase.

If interrupted, the victim returns from its yield hold statement prematurely. It must then check to see if it has beeninterrupted by calling:

interrupted

• self.interrupted( )

which returns True if it has been interrupted. The victim can then either continue in the current activity orswitch to an alternative, making sure it tidies up the current state, such as releasing any resources it owns.

interruptCause

• self.interruptCause

when the victim has been interrupted, self.interruptCause is a reference to the interrupter object.

interruptLeft

• self.interruptLeft

gives the time remaining in the interrupted yield hold. The interruption is reset (that is, “turned off”) at thevictim’s next call to a yield hold.

..index:: interruptReset

interruptReset

• self.interruptReset( )

will reset the interruption.

It may be helpful to think of an interruption signal as instructing the victim to determine whether it should interruptitself. If the victim determines that it should interrupt itself, it then becomes responsible for making any necessaryreadjustments – not only to itself but also to any other simulation components that are affected. (The victim must takeresponsibility for these adjustments, because it is the only simulation component that “knows” such details as whetheror not it is interrupting itself, when, and why.)



Example 5. A simulation with interrupts. A bus is subject to breakdowns that are modelled as interrupts caused bya Breakdown process. Notice that the yield hold,self,tripleft statement may be interrupted, so if theself.interrupted() test returns True a reaction to it is required. Here, in addition to delaying the bus forrepairs, the reaction includes scheduling the next breakdown. In this example the Bus Process class does not requirean __init__() method:

from SimPy.Simulation import (Process, activate, hold, initialize, now,reactivate, simulate)

class Bus(Process):

def operate(self, repairduration, triplength): # PEMtripleft = triplength # time needed to finish tripwhile tripleft > 0:

yield hold, self, tripleft # try to finish the tripif self.interrupted(): # if another breakdown occurs

print('%s at %s' % (self.interruptCause.name, now()))tripleft = self.interruptLeft # time to finish the tripself.interruptReset() # end interrupt statereactivate(br, delay=repairduration) # restart breakdown bryield hold, self, repairduration # delay for repairsprint('Bus repaired at %s' % now())

else:break # no more breakdowns, bus finished trip

print('Bus has arrived at %s' % now())

class Breakdown(Process):def __init__(self, myBus):

Process.__init__(self, name='Breakdown ' + myBus.name)self.bus = myBus

def breakBus(self, interval): # process execution methodwhile True:

yield hold, self, interval # breakdown interarrivalsif self.bus.terminated():

breakself.interrupt(self.bus) # breakdown to myBus

initialize()b = Bus('Bus') # create a bus objectactivate(b, b.operate(repairduration=20, triplength=1000))br = Breakdown(b) # create breakdown br to bus bactivate(br, br.breakBus(300))simulate(until=4000)print('SimPy: No more events at time %s' % now())

The output from this example:

Breakdown Bus at 300Bus repaired at 320Breakdown Bus at 620Bus repaired at 640Breakdown Bus at 940Bus repaired at 960Bus has arrived at 1060



SimPy: No more events at time 1260

The bus finishes at 1060 but the simulation finished at 1260. Why? The breakdowns PEM consists of a loop, onebreakdown following another at 300 intervals. The last breakdown finishes at 960 and then a breakdown event isscheduled for 1260. But the bus finished at 1060 and is not affected by the breakdown. These details can easily bechecked by importing from SimPy.SimulationTrace and re-running the program.

Where interrupts can occur, the victim of interrupts must test for interrupt occurrence after every appropriate yieldhold and react appropriately to it. A victim holding a resource facility when it gets interrupted continues to hold it.

Advanced synchronization/scheduling capabilities

The preceding scheduling constructs all depend on specified time values. That is, they delay processes for a specifictime, or use given time parameters when reactivating them. For a wide range of applications this is all that is needed.

However, some applications either require or can profit from an ability to activate processes that must wait for otherprocesses to complete. For example, models of real-time systems or operating systems often use this kind of approach.Event Signalling is particularly helpful in such situations. Furthermore, some applications need to activate processeswhen certain conditions occur, even though when (or if) they will occur may be unknown. SimPy has a general waituntil to support clean implementation of this approach.

This section describes how SimPy provides event Signalling and wait until capabilities.

Creating and Signalling SimEvents

As mentioned in the Introduction, for ease of expression when no confusion can arise we often refer to both processobjects and their classes as “processes”, and mention their object or class status only for added clarity or emphasis.Analogously, we will refer to objects of SimPy’s SimEvent class as “SimEvents”3 (or, if no confusion can arise,simply as “events”). However, we sometimes mention their object or class character for clarity or emphasis.

SimEvent objects must be created before they can be fired by a signal. You create the SimEvent object, sE, fromSimPy’s SimEvent class by a statement like the following:

sE = SimEvent(name='I just had a great new idea!')

A SimEvent’s name attribute defaults to a_SimEvent unless you provide your own, as shown here. Its occurredattribute, sE.occurred, is a Boolean that defaults to False. It indicates whether the event sE has occurred.

You program a SimEvent to “occur” or “fire” by “signalling” it like this:

sE.signal(<payload parameter>)

This “signal” is “received” by all processes that are either “waiting” or “queueing” for this event to occur. Whathappens when they receive this signal is explained in the next section. The <payload parameter> is optional – itdefaults to None. It can be of any Python type. Any process can retrieve it from the event’s signalparam attribute,for example by:

message = sE.signalparam

3 unless it is further delayed by being interrupted.This is used to model any elapsed time an entity might be involved in. For example while it is passively being provided with service.



Waiting or Queueing for SimEvents

You can program a process either to “wait” or to “queue” for the occurrence of SimEvents. The difference is that allprocesses “waiting” for some event are reactivated as soon as it occurs. For example, all firemen go into action whenthe alarm sounds. In contrast, only the first process in the “queue” for some event is reactivated when it occurs. Thatis, the “queue” is FIFO5. An example might be royal succession – when the present ruler dies: “The king is dead.Long live the (new) king!” (And all others in the line of succession move up one step.)

You program a process to wait for SimEvents by including in its PEM:

yield waitevent

• yield waitevent, self,<events part>

where <events part> can be either:

– one SimEvent object, e.g. myEvent, or

– a tuple of SimEvent objects, e.g. (myEvent,myOtherEvent,TimeOut), or

– a list of SimEvent objects, e.g. [myEvent,myOtherEvent,TimeOut]

If none of the events in the <events part> have occurred, the process is passivated and joined to the list ofprocesses waiting for some event in <events part> to occur (or to recur).

On the other hand, when any of the events in the <events part> occur, then all of the processes “waiting”for those particular events are reactivated at the current time. Then the occurred flag of those particularevents is reset to False. Resetting their occurred flag prevents the waiting processes from being constantlyreactivated. (For instance, we do not want firemen to keep responding to any such “false alarms.”) For example,suppose the <events part> lists events a, b and c in that order. If events a and c occur, then all of the processeswaiting for event a are reactivated. So are all processes waiting for event c but not a. Then the occurredflags of events a and c are toggled to False. No direct changes are made to event b or to any processes waitingfor it to occur.

You program a process to “queue” for events by including in its PEM:

yield queueevent

• yield queueevent, self,<events part>

where the <events part> is as described above.

If none of the events in the <events part> has occurred, the process is passivated and appended to the FIFOqueue of processes queuing for some event in <events part> to occur (or recur).

But when any of the events in <events part> occur, the process at the head of the “queue” is taken off the queueand reactivated at the current time. Then the occurred flag of those events that occurred is reset to False asin the “waiting” case.

Finding Which Processes Are Waiting/Queueing for an Event, and Which Events Fired

SimPy automatically keeps current lists of what processes are “waiting” or “queueing” for SimEvents. They are keptin the waits and queues attributes of the SimEvent object and can be read by commands like the following:

5 “First-in-First-Out” or FCFS, “First-Come-First-Served”



TheProcessesWaitingFor_myEvent = myEvent.waitsTheProcessesQueuedFor_myEvent = myEvent.queues

However, you should not attempt to change these attributes yourself.

Whenever myEvent occurs, i.e., whenever a myEvent.signal(...) statement is executed, SimPy does thefollowing:

• If there are any processes waiting or queued for that event, it reactivates them as described in the precedingsection.

• If there are no processes waiting or queued (i.e., myEvent.waits and myEvent.queues are both empty),it toggles myEvent.occurred to True.

SimPy also automatically keeps track of which events were fired when a process object was reactivated. For example,you can get a list of the events that were fired when the object Godzilla was reactivated with a statement like this:

GodzillaRevivedBy = Godzilla.eventsFired

Example 6. This complete SimPy script illustrates these constructs. (It also illustrates that a Process class may havemore than one PEM. Here the Wait_Or_Queue class has two PEMs – waitup and queueup.):


class Wait_Or_Queue(Process):def waitup(self, myEvent): # PEM illustrating "waitevent"

# wait for "myEvent" to occuryield waitevent, self, myEventprint('At %s, some SimEvent(s) occurred that activated object %s.' %

(now(), self.name))print(' The activating event(s) were %s' %

([x.name for x in self.eventsFired]))

def queueup(self, myEvent): # PEM illustrating "queueevent"# queue up for "myEvent" to occur

yield queueevent, self, myEventprint('At %s, some SimEvent(s) occurred that activated object %s.' %

(now(), self.name))print(' The activating event(s) were %s' %

([x.name for x in self.eventsFired]))

class Signaller(Process):# here we just schedule some events to firedef sendSignals(self):

yield hold, self, 2event1.signal() # fire "event1" at time 2yield hold, self, 8event2.signal() # fire "event2" at time 10yield hold, self, 5event1.signal() # fire all four events at time 15event2.signal()event3.signal()event4.signal()yield hold, self, 5event4.signal() # event4 recurs at time 20



initialize()

# Now create each SimEvent and give it a nameevent1 = SimEvent('Event-1')event2 = SimEvent('Event-2')event3 = SimEvent('Event-3')event4 = SimEvent('Event-4')Event_list = [event3, event4] # define an event list

s = Signaller()# Activate Signaller "s" *after* events createdactivate(s, s.sendSignals())

w0 = Wait_Or_Queue('W-0')# create object named "W-0", and set it to# "waitup" for SimEvent "event1" to occuractivate(w0, w0.waitup(event1))w1 = Wait_Or_Queue('W-1')activate(w1, w1.waitup(event2))w2 = Wait_Or_Queue('W-2')activate(w2, w2.waitup(Event_list))q1 = Wait_Or_Queue('Q-1')# create object named "Q-1", and put it to be first# in the queue for Event_list to occuractivate(q1, q1.queueup(Event_list))q2 = Wait_Or_Queue('Q-2')# create object named "Q-2", and append it to# the queue for Event_list to occuractivate(q2, q2.queueup(Event_list))

simulate(until=50)

This program outputs:

At 2, some SimEvent(s) occurred that activated object W-0.The activating event(s) were ['Event-1']



At 15, some SimEvent(s) occurred that activated object Q-1.The activating event(s) were ['Event-3', 'Event-4']

At 20, some SimEvent(s) occurred that activated object Q-2.The activating event(s) were ['Event-4']

Each output line, The activating event(s) were ..., lists the contents of the named object’seventsFired attribute. One of those events “caused” the object to reactivate at the indicated time. Note that attime 15 objects W-0 and W-1 were not affected by the recurrence of event1 and event2 because they already wereactive. Also at time 15, even though objects W-2, Q-1 and Q-2 were all waiting for event3, only W-2 and Q-1were reactivated. Process object Q-2 was not reactivated at that time because it was not first in the queue. Finally,Q-2 was reactivated at time 20, when event4 fired again.



“waituntil” synchronization – waiting for any condition

SimPy provides the waituntil feature that makes a process’s progress depend on the state of the simulation. This isuseful if, for example, you need to reactivate a process when (if ever) the simulation enters the state goodWeatherOR (nrCustomers>50 AND price<22.50). Doing that requires interrogative scheduling, while all otherSimPy synchronization constructs are imperative – i.e., the condition must be tested after every change in state until itbecomes True.

This requires that after every change in system state SimPy must run a special (hidden) process that tests and respondsappropriately to the condition’s truth-value. This clearly takes more run time than SimPy’s imperative schedulingconstructs. So SimPy activates its interrogative testing process only so long as at least one process is executing awaituntil statement. When this is not the case, the run time overhead is minimal (about 1 percent extra run time).

yield waituntil

You program a process to wait for a condition to be satisfied by including in its PEM a statement of the form:

yield waituntil, self,<cond>

where <cond> is a reference to a function, without parameters, that returns a Boolean value indicating whether thesimulation state or condition to be waited for has occurred.

Example 7. This program using the yield waituntil ... statement. Here the function killed(), in thelife() PEM of the Player process, defines the condition to be waited for:

from SimPy.Simulation import (Process, initialize, activate, simulate,hold, now, waituntil, stopSimulation)

import random

class Player(Process):

def __init__(self, lives=1, name='ImaTarget'):Process.__init__(self, name)self.lives = lives# provide Player objects with a "damage" propertyself.damage = 0

def life(self):self.message = 'Drat! Some %s survived Federation attack!' % \

(target.name)

def killed(): # function testing for "damage > 5"return self.damage > 5

while True:yield waituntil, self, killedself.lives -= 1self.damage = 0if self.lives == 0:

self.message = '%s wiped out by Federation at \time %s!' % (target.name, now())stopSimulation()

class Federation(Process):



def fight(self): # simulate Federation operationsprint('Three %s attempting to escape!' % (target.name))while True:

if random.randint(0, 10) < 2: # check for hit on playertarget.damage += 1 # hit! increment damage to playerif target.damage <= 5: # target survives

print('Ha! %s hit! Damage= %i' %(target.name, target.damage))

else:if (target.lives - 1) == 0:

print('No more %s left!' % (target.name))else:

print('Now only %i %s left!' % (target.lives - 1,target.name))

yield hold, self, 1

initialize()gameOver = 100# create a Player object named "Romulans"target = Player(lives=3, name='Romulans')activate(target, target.life())# create a Federation objectshooter = Federation()activate(shooter, shooter.fight())simulate(until=gameOver)print(target.message)

One possible output from this program is shown below. Whether the Romulans are wiped out or some escape dependson what simulation states the randomization feature produces:

Three Romulans attempting to escape!Ha! Romulans hit! Damage= 1Ha! Romulans hit! Damage= 2Ha! Romulans hit! Damage= 3Ha! Romulans hit! Damage= 4Ha! Romulans hit! Damage= 5Now only 2 Romulans left!Ha! Romulans hit! Damage= 1Ha! Romulans hit! Damage= 2Ha! Romulans hit! Damage= 3Ha! Romulans hit! Damage= 4Ha! Romulans hit! Damage= 5Now only 1 Romulans left!Ha! Romulans hit! Damage= 1Ha! Romulans hit! Damage= 2Drat! Some Romulans survived Federation attack!

The waituntil construct is so general that in principle it could replace all the other synchronization approaches(but at a run time cost).

[Return to Top ]



2.1.4 Resources

The three resource facilities provided by SimPy are Resources, Levels and Stores. Each models a congestion pointwhere process objects may have to queue up to obtain resources. This section describes the Resource type of resourcefacility.

An example of queueing for a Resource might be a manufacturing plant in which a Task (modelled as a processobject) needs work done by a Machine (modelled as a Resource object). If all of the Machines are currently beingused, the Task must wait until one becomes free. A SimPy Resource can have a number of identical units, such asa number of identical machine units. A process obtains a unit of the Resource by requesting it and, when it isfinished, releasing it. A Resource maintains a list of process objects that have requested but not yet received one ofthe Resource’s units (called the waitQ), and another list of processes that are currently using a unit (the activeQ).SimPy creates and updates these queues itself – the user can access them, but should not change them.

Defining a Resource object

A Resource object, r, is established by the following statement:

r = Resource(capacity=1, name='a_resource', unitName='units',qType=FIFO, preemptable=False,monitored=False, monitorType=Monitor)

where

• capacity is a positive real or integer value that specifies the total number of identical units in Resource objectr.

• name is a descriptive name for this Resource object (e.g., 'gasStation').

• unitName is a descriptive name for a unit of the resource (e.g., 'pump').

• qType is either FIFO5 or PriorityQ. It specifies the queue discipline of the resource’s waitQ; typically,this is FIFO and that is the default value. If PriorityQ is specified, then higher-priority requests waiting fora unit of Resource r are inserted into the waitQ ahead of lower priority requests. See Priority requests for aResource unit for details.

• preemptable is a Boolean (False or True); typically, this is False and that is the default value. If it isTrue, then a process requesting a unit of this resource may preempt a lower-priority process in the activeQ,i.e., one that is already using a unit of the resource. See Preemptive requests for a Resource unit for details.

• monitored is a boolean (False or True). If set to True, then information is gathered on the sizes of r’swaitQ and activeQ, otherwise not.

• monitorType is either Monitor or Tally and indicates the type of Recorder to be used (see RecordingResource queue lengths for an example and additional discussion).

Each Resource object, r, has the following additional attributes:

• r.n, the number of units that are currently free.

• r.waitQ, a queue (list) of processes that have requested but not yet received a unit of r, so len(r.waitQ)is the number of process objects currently waiting.

• r.activeQ, a queue (list) of process objects currently using one of the Resource’s units, so len(r.activeQ) is the number of units that are currently in use.

• r.waitMon, the record (made by a Monitor or a Tally whenever monitored==True) of the activityin r.waitQ. So, for example, r.waitMon.timeaverage() is the average number of processes in r.waitQ. See Recording Resource queue lengths for an example.



• r.actMon, the record (made by a Monitor or a Tally whenever monitored==True) of the activity inr.activeQ.

Requesting and releasing a unit of a Resource

A process can request and later release a unit of the Resource object, r, by using the following yield commands in aProcess Execution Method:

yield request

• yield request,self,r [,P=0]

requests a unit of Resource r with (optional) real or integer priority value P. If no priority is specified, it defaultsto 0. Larger values of P represent higher priorities. See the following sections on Queue Order for moreinformation on how priority values are used. Although this form of request can be used for either FIFO orPriorityQ priority types, these values are ignored when qType==FIFO.

yield release

yield release, self, r

releases the unit of r.

Queue Order

If a requesting process must wait it is placed into the resource’s waitQ in an order determined by settings of theresource’s qType and preemptable attributes and of the priority value it uses in the request call.

Non-priority queueing

If the qType is not specified it takes the presumed value of FIFO5. In that case processes wait in the usual first-come-first-served order.

If a Resource unit is free when the request is made, the requesting process takes it and moves on to the next statementin its PEM. If no Resource unit is available when the request is made, then the requesting process is appended tothe Resource’s waitQ and suspended. The next time a unit becomes available the first process in the r.waitQtakes it and continues its execution. All priority assignments are ignored. Moreover, in the FIFO case no preemptionis possible, for preemption requires that priority assignments be recognized. (However, see the Note on preemptiverequests with waitQ in FIFO order for one way of simulating such situations.)

Example In this complete script, the server Resource object is given two resource units (capacity=2). By notspecifying its Qtype it takes the default value, FIFO. Here six clients arrive in the order specified by the program.They all request a resource unit from the server Resource object at the same time. Even though they all specify apriority value in their requests, it is ignored and they get their Resource units in the same order as their requests:

from SimPy.Simulation import (Process, Resource, activate, initialize, hold,now, release, request, simulate)

class Client(Process):



inClients = [] # list the clients in order by their requestsoutClients = [] # list the clients in order by completion of service

def __init__(self, name):Process.__init__(self, name)

def getserved(self, servtime, priority, myServer):Client.inClients.append(self.name)print('%s requests 1 unit at t = %s' % (self.name, now()))# request use of a resource unityield request, self, myServer, priorityyield hold, self, servtime# release the resourceyield release, self, myServerprint('%s done at t = %s' % (self.name, now()))Client.outClients.append(self.name)

initialize()

# the next line creates the ``server`` Resource objectserver = Resource(capacity=2) # server defaults to qType==FIFO

# the next lines create some Client process objectsc1, c2 = Client(name='c1'), Client(name='c2')c3, c4 = Client(name='c3'), Client(name='c4')c5, c6 = Client(name='c5'), Client(name='c6')

# in the next lines each client requests# one of the ``server``'s Resource unitsactivate(c1, c1.getserved(servtime=100, priority=1, myServer=server))activate(c2, c2.getserved(servtime=100, priority=2, myServer=server))activate(c3, c3.getserved(servtime=100, priority=3, myServer=server))activate(c4, c4.getserved(servtime=100, priority=4, myServer=server))activate(c5, c5.getserved(servtime=100, priority=5, myServer=server))activate(c6, c6.getserved(servtime=100, priority=6, myServer=server))

simulate(until=500)

print('Request order: %s' % Client.inClients)print('Service order: %s' % Client.outClients)

This program results in the following output:

1 c1 requests 1 unit at t = 02 c2 requests 1 unit at t = 03 c3 requests 1 unit at t = 04 c4 requests 1 unit at t = 05 c5 requests 1 unit at t = 06 c6 requests 1 unit at t = 07 c1 done at time = 1008 c2 done at time = 1009 c3 done at time = 200

10 c4 done at time = 20011 c5 done at time = 30012 c6 done at time = 30013

14 Request order: ['c1', 'c2', 'c3', 'c4', 'c5', 'c6']



15 Service order: ['c1', 'c2', 'c3', 'c4', 'c5', 'c6']

As illustrated, the clients are served in FIFO order. Clients c1 and c2 each take one Resource unit right away, but theothers must wait. When c1 and c2 finish with their resources, clients c3 and c4 can each take a unit, and so forth.

Priority requests for a Resource unit

If the Resource r is defined with qType==PriorityQ, priority values in requests are recognized. If a Resourceunit is available when the request is made, the requesting process takes it. If no Resource unit is available when therequest is made, the requesting process is inserted into the Resource’s waitQ in order of priority (from high to low)and suspended. For an example where priorities are used, we simply change the preceding example’s specification ofthe server Resource object to:

server = Resource(capacity=2, qType=PriorityQ)

where, by not specifying it, we allow preemptable to take its default value, False.

Example After this change the program’s output becomes:

1 c1 requests 1 unit at t = 02 c2 requests 1 unit at t = 03 c3 requests 1 unit at t = 04 c4 requests 1 unit at t = 05 c5 requests 1 unit at t = 06 c6 requests 1 unit at t = 07 c1 done at time = 1008 c2 done at time = 1009 c6 done at time = 200

10 c5 done at time = 20011 c4 done at time = 30012 c3 done at time = 30013

14 Request order: ['c1', 'c2', 'c3', 'c4', 'c5', 'c6']15 Service order: ['c1', 'c2', 'c6', 'c5', 'c4', 'c3']

Although c1 and c2 have the lowest priority values, each requested and got a server unit immediately. That wasbecause at the time they made those requests a server unit was available and the server.waitQ was empty – itdid not start to fill until c3 made its request and found all of the server units busy. When c1 and c2 completedservice, c6 and c5 (with the highest priority values of all processes in the waitQ) each got a Resource unit, etc.

When some processes in the waitQ have the same priority level as a process making a priority request, SimPyinserts the requesting process immediately behind them. Thus for a given priority value, processes are placed inFIFO order. For example, suppose that when a “priority 3” process makes its priority request the current waitQconsists of processes with priorities [5,4,3a,3b,3c,2a,2b,1], where the letters indicate the order in which theequal-priority processes were placed in the queue. Then SimPy inserts this requesting process into the current waitQimmediately behind its last “priority 3” process. Thus, the new waitQ will be [5,4,3a,3b,3c,3d,2a,2b,1],where the inserted process is 3d.

One consequence of this is that, if all priority requests are assigned the same priority value, then the waitQ will in factbe maintained in FIFO order. In that case, using a FIFO instead of a PriorityQ discipline provides some savingin execution time which may be important in simulations where the waitQ may be long.



Preemptive requests for a Resource unit

In some models, higher priority processes can actually preempt lower priority processes, i.e., they can take over anduse a Resource unit currently being used by a lower priority process whenever no free Resource units are available. AResource object that allows its units to be preempted is created by setting its properties to qType == PriorityQand preemptable == True.

Whenever a preemptable Resource unit is free when a request is made, then the requesting process takes it andcontinues its execution. On the other hand, when a higher priority request finds all the units in a preemptableResource in use, then SimPy adopts the following procedure regarding the Resource’s activeQ and waitQ:

• The process with the lowest priority is removed from the activeQ, suspended, and put at the front of thewaitQ – so (barring additional preemptions) it will be the next one to get a resource unit.

• The preempting process gets the vacated resource unit and is inserted into the activeQ in order of its priorityvalue.

• The time for which the preempted process had the resource unit is taken into account when the process gets intothe activeQ again. Thus, its total hold time is always the same, regardless of how many times it has beenpreempted.

Warning: SimPy only supports preemption of processes which are implemented in the following pattern:

1 yield request (one or more request statements)2 <some code>3 yield hold (one or more hold statements)4 <some code>5 yield release (one or more release statements)

Modelling the preemption of a process in any other pattern may lead to errors or exceptions.

We emphasize that a process making a preemptive request to a fully-occupied Resource gets a resource unit if –but only if – some process in the current activeQ has a lower priority. Otherwise, it will be inserted into the waitQat a location determined by its priority value and the current contents of the waitQ, using a procedure analogous tothat described for priority requests near the end of the preceding section on Priority requests for a Resource unit. Thismay have the effect of advancing the preempting process ahead of any lower-priority processes that had earlier beenpreempted and put at the head of the waitQ. In fact, if several preemptions occur before a unit of resource is freedup, then the head of the waitQ will consist of the processes that have been preempted – in order from the last processpreempted to the first of them.

Example In this example two clients of different priority compete for the same resource unit:

from SimPy.Simulation import (PriorityQ, Process, Resource, activate,initialize, hold, now, release, request,simulate)

class Client(Process):def __init__(self, name):

Process.__init__(self, name)

def getserved(self, servtime, priority, myServer):print('%s requests 1 unit at t=%s' % (self.name, now()))yield request, self, myServer, priorityyield hold, self, servtimeyield release, self, myServerprint('%s done at t=%s' % (self.name, now()))



initialize()# create the *server* Resource objectserver = Resource(capacity=1, qType=PriorityQ, preemptable=1)# create some Client process objectsc1 = Client(name='c1')c2 = Client(name='c2')activate(c1, c1.getserved(servtime=100, priority=1, myServer=server), at=0)activate(c2, c2.getserved(servtime=100, priority=9, myServer=server), at=50)simulate(until=500)

The output from this program is:

c1 requests 1 unit at t=0c2 requests 1 unit at t=50c2 done at t=150c1 done at t=200

Here, c1 is preempted by c2 at t=50. At that time, c1 had held the resource for 50 of its total of 100 time units.When c2 finished and released the resource unit at 150, c1 got the resource back and finished the last 50 time unitsof its service at t=200.

If preemption occurs when the last few processes in the current activeQ have the same priority value, then the lastprocess in the current activeQ is the one that will be preempted and inserted into the waitQ ahead of all others.To describe this, it will be convenient to indicate by an added letter the order in which equal-priority processes havebeen inserted into a queue. Now, suppose that a “priority 4” process makes a preemptive request when the currentactiveQ priorities are [5,3a,3b] and the current waitQ priorities are [2,1,0a,0b]. Then process 3b will bepreempted. After the preemption the activeQ will be [5,4,3a] and the waitQ will be [3b,2,1,0a,0b].

Note on preemptive requests with waitQ in FIFO order

You may consider doing the following to model a system whose queue of items waiting for a resource is to be main-tained in FIFO order, but in which preemption is to be possible. It uses SimPy’s preemptable Resource objects,and uses priorities in a way that allows for preempts while maintaining a FIFO waitQ order.

• Set qType=PriorityQ and preemptable=True (so that SimPy will process preemptive requests cor-rectly).

• Model “system requests that are to be considered as non-preemptive” in SimPy as process objects each of whichhas exactly the same (low) priority value – for example, either assign all of them a priority value of 0 (zero) orlet it default to that value. (This has the effect of maintaining all of these process objects in the waitQ in FIFOorder, as explained at the end of the section on Priority requests for a Resource unit, above.)

• Model “system requests that are to be considered as preemptive” in SimPy as process objects each of which isassigned a uniform priority value, but give them a higher value than the one used to model the “non-preemptivesystem requests” – for example, assign all of them a priority value of 1 (one). Then they will have a higherpriority value than any of the non-preemptive requests.

Example Here is an example of how this works for a Resource with two Resource units – we give the activeQbefore the waitQ throughout this example:

1. Suppose that the current activeQ and waitQ are [0a,0b] and [0c], respectively.



2. A “priority 1” process makes a preemptive request. Then the queues become: [1a,0a] and‘‘ [0b,0c]‘‘.

3. Another “priority 1” process makes a preemptive request. Then the queues become: [1a,1b] and [0a,0b,0c].

4. A third “priority 1” process makes a preemptive request. Then the queues become: [1a,1b] and [1c,0a,0b,0c].

5. Process 1a finishes using its resource unit. Then the queues become: [1b,1c] and [0a,0b,0c].

Reneging – leaving a queue before acquiring a resource

In most real world situations, people and other items do not wait forever for a requested resource facility to becomeavailable. Instead, they leave its queue when their patience is exhausted or when some other condition occurs. Thisbehaviour is called reneging, and the reneging person or thing is said to renege.

SimPy provides an extended (i.e., compound) yield request statement to handle reneging.

Reneging yield request

There are two types of reneging clause, one for reneging after a certain time and one for reneging when an event hashappened. Their general form is

yield (request,self,r [,P]),(<reneging clause>)

to request a unit of Resource r (with optional priority P, assuming the Resource has been defined as a priorityQ)but with reneging.

A SimPy program that models Resource requests with reneging must use the following pattern of statements:

1 yield (request, self, r), (<reneging clause>)2 if self.acquired(resource):3 ## process got resource and so did NOT renege4 . . . .5 yield release, self, resource6 else:7 ## process reneged before acquiring resource8 . . . . .

A call to the self.acquired(resource) method is mandatory after a compound yield request statement.It not only indicates whether or not the process has acquired the resource, it also removes the reneging process fromthe resource’s waitQ.

Reneging after a time limit

To make a process give up (renege) after a certain time, use a reneging clause of the following form:

yield (request,self,r [,P]),(hold, self, ``\ *waittime*\ ``)

Here the process requests one unit of the resource r with optional priority P. If a resource unit is available it takes itand continues its PEM. Otherwise, as usual, it is passivated and inserted into r’s waitQ.

The process takes a unit if it becomes available before waittime expires and continues executing its PEM. If, however,the process has not acquired a unit before the waittime has expired it abandons the request (reneges) and leaves thewaitQ.



Example: part of a parking lot simulation:

1 . . . .2 parking_lot = Resource(capacity=10)3 patience = 5 # wait no longer than "patience" time units4 # for a parking space5 park_time = 60 # park for "park_time" time units if get a parking space6 . . . .7 yield (request, self, parking_lot), (hold, self, patience)8 if self.acquired(parking_lot):9 # park the car

10 yield hold, self, park_time11 yield release, self, parking_lot12 else:13 # patience exhausted, so give up14 print("I'm not waiting any longer. I am going home now.")

Reneging when an event has happened

To make a process renege at the occurrence of an event, use a reneging clause having a pattern like the one used for ayield waitevent statement, namely waitevent,self,events (see yield waitevent). For example:

yield (request,self,r [,P]),(waitevent,self,events)

Here the process requests one unit of the resource r with optional priority P. If a resource unit is available it takes itand continues its PEM. Otherwise, as usual, it is passivated and inserted into r’s waitQ.

The process takes a unit if it becomes available before any of the events occur, and continues executing its PEM. If,however, any of the SimEvents in events occur first, it abandons the request (reneges) and leaves the waitQ. (Recallthat events can be either one event, a list, or a tuple of several SimEvents.)

Example Queuing for movie tickets (part):

1 . . . .2 seats = Resource(capacity=100)3 sold_out = SimEvent() # signals "out of seats"4 too_late = SimEvent() # signals "too late for this show"5 . . . .6 # Leave the ticket counter queue when movie sold out7 # or it is too late for the show8 yield (request, self, seats), (waitevent, self, [sold_out, too_late])9 if self.acquired(seats):

10 # watch the movie11 yield hold, self, 12012 yield release, self, seats13 else:14 # did not get a seat15 print('Who needs to see this silly movie anyhow?')



Exiting conventions and preemptive queues

Many discrete event simulations (including SimPy) adopt the normal “exiting convention”, according to which pro-cesses that have once started using a Resource unit stay in some Resource queue until their hold time has completed.This is of course automatically the case for FIFO and non-preemptable PriorityQ disciplines. The point is thatthe exiting convention is also applied in the preemptable queue discipline case. Thus, processes remain in someResource queue until their hold time has completed, even if they are preempted by higher priority processes.

Some real-world situations conform to this convention and some do not. An example of one that does conform canbe described as follows. Suppose that at work you are assigned tasks of varying levels of priority. You are to setaside lower priority tasks in order to work on higher priority ones. But you are eventually to complete all of yourassigned tasks. So you are operating like a SimPy resource that obeys a preemptable queue discipline and hasone resource unit. With this convention, half-finished low-priority tasks may be postponed indefinitely if they arecontinually preempted by higher-priority tasks.

An example that does not conform to the exiting convention can be described as follows. Suppose again that you areassigned tasks of varying levels of priority and are to set aside lower priority tasks to work on higher priority ones.But you are instructed that any tasks not completed within 24 hours after being assigned are to be sent to anotherdepartment for completion. Now, suppose that you are assigned Task-A that has a priority level of 3 and will take 10hours to complete. After working on Task-A for an hour, you are assigned Task-B, which has a priority level of 5 andwill take 20 hours to complete. Then, at 11 hours, after working on Task-B for 10 hours, you are assigned Task-C,which has a priority level of 1 and will take 4 hours to complete. (At this point Task-B needs 10 hours to complete,Task-A needs 9 hours to complete, and Task-C needs 4 hours to complete.) At 21 hours you complete Task-B andresume working on Task-A, which at that point needs 9 hours to complete. At 24 hours Task-A still needs another 6hours to complete, but it has reached the 24-hour deadline and so is sent to another department for completion. Atthe same time, Task-C has been in the waitQ for 13 hours, so you take it up and complete it at hour 28. This queuediscipline does not conform to the exiting convention, for under that convention at 24 hours you would continue workon Task-A, complete it at hour 30, and then start on Task-C.

Recording Resource queue lengths

Many discrete event models are used mainly to explore the statistical properties of the waitQ and activeQ associ-ated with some or all of their simulated resources. SimPy’s support for this includes the Monitor and the Tally. Formore information on these and other recording methods, see the section on Recording Simulation Results.

If a Resource, r, is defined with monitored=True SimPy automatically records the length of its associated waitQand activeQ. These records are kept in the recorder objects called r.waitMon and r.actMon, respectively. Thissolves a problem, particularly for the waitQ which cannot easily be recorded externally to the resource.

The property monitorType indicates which variety of recorder is to be used, either Monitor or Tally. The default isMonitor. If this is chosen, complete time series for both queue lengths are maintained and can be used for advancedpost-simulation statistical analyses as well as for displaying summary statistics (such as averages, standard deviations,and histograms). If Tally is chosen summary statistics can be displayed, but complete time series cannot. For moreinformation on these and SimPy’s other recording methods, see the section on Recording Simulation Results.

Example The following program uses a Monitor to record the server resource’s queues. After the simulationends, it displays some summary statistics for each queue, and then their complete time series:

from math import sqrtfrom SimPy.Simulation import (Monitor, Process, Resource, activate, initialize,

hold, now, release, request, simulate)

class Client(Process):



inClients = []outClients = []


def getserved(self, servtime, myServer):print('%s requests 1 unit at t = %s' % (self.name, now()))yield request, self, myServeryield hold, self, servtimeyield release, self, myServerprint('%s done at t = %s' % (self.name, now()))

initialize()

server = Resource(capacity=1, monitored=True, monitorType=Monitor)

c1, c2 = Client(name='c1'), Client(name='c2')c3, c4 = Client(name='c3'), Client(name='c4')

activate(c1, c1.getserved(servtime=100, myServer=server))activate(c2, c2.getserved(servtime=100, myServer=server))activate(c3, c3.getserved(servtime=100, myServer=server))activate(c4, c4.getserved(servtime=100, myServer=server))

simulate(until=500)

print('')print('(TimeAverage no. waiting: %s' % server.waitMon.timeAverage())print('(Number) Average no. waiting: %.4f' % server.waitMon.mean())print('(Number) Var of no. waiting: %.4f' % server.waitMon.var())print('(Number) SD of no. waiting: %.4f' % sqrt(server.waitMon.var()))print('(TimeAverage no. in service: %s' % server.actMon.timeAverage())print('(Number) Average no. in service: %.4f' % server.actMon.mean())print('(Number) Var of no. in service: %.4f' % server.actMon.var())print('(Number) SD of no. in service: %.4f' % sqrt(server.actMon.var()))print('=' * 40)print('Time history for the "server" waitQ:')print('[time, waitQ]')for item in server.waitMon:

print(item)print('=' * 40)print('Time history for the "server" activeQ:')print('[time, activeQ]')for item in server.actMon:

print(item)

The output from this program is:

c1 requests 1 unit at t = 0c2 requests 1 unit at t = 0c3 requests 1 unit at t = 0c4 requests 1 unit at t = 0c1 done at t = 100c2 done at t = 200c3 done at t = 300c4 done at t = 400



(TimeAverage no. waiting: 1.5(Number) Average no. waiting: 1.2857(Number) Var of no. waiting: 1.0612(Number) SD of no. waiting: 1.0302(TimeAverage no. in service: 1.0(Number) Average no. in service: 0.4444(Number) Var of no. in service: 0.2469(Number) SD of no. in service: 0.4969========================================Time history for the "server" waitQ:[time, waitQ][0, 0][0, 1][0, 2][0, 3][100, 2][200, 1][300, 0]========================================Time history for the "server" activeQ:[time, activeQ][0, 0][0, 1][100, 0][100, 1][200, 0][200, 1][300, 0][300, 1][400, 0]

This output illustrates the difference between the (Time) Average and the number statistics. Here process c1 was inthe waitQ for zero time units, process c2 for 100 time units, and so forth. The total wait time accumulated by all fourprocesses during the entire simulation run, which ended at time 400, amounts to 0 + 100 + 200 + 300 = 600 time units.Dividing the 600 accumulated time units by the simulation run time of 400 gives 1.5 for the (Time) Average numberof processes in the waitQ. It is the time-weighted average length of the waitQ, but is almost always called simplythe average length of the waitQ or the average number of items waiting for a resource.

It is also the expected number of processes you would find in the waitQ if you took a snapshot of it at a random timeduring the simulation. The activeQ’s time average computation is similar, although in this example the resourceis held by some process throughout the simulation. Even though the number in the activeQ momentarily dropsto zero as one process releases the resource and immediately rises to one as the next process acquires it, that occursinstantaneously and so contributes nothing to the (Time) Average computation.

Number statistics such as the Average, Variance, and Standard Deviation are computed differently. At time zero thenumber of processes in the waitQ starts at 1, then rises to 2, and then to 3. At time 100 it drops back to two processes,and so forth. The average and standard deviation of the six values [1, 2, 3, 2, 1, 0] is 1.5 and 0.9574. . . , respectively.Number statistics for the activeQ are computed using the eight values [1, 0, 1, 0, 1, 0, 1, 0] and are as shown in theoutput.

When the monitorType is changed to Tally, all the output up to and including the lines:

Time history for the 'server' waitQ:[time, waitQ]



is displayed. Then the output concludes with an error message indicating a problem with the reference to server.waitMon. Of course, this is because Tally does not generate complete time series.

[Return to Top ]

2.1.5 Levels

The three resource facilities provided by the SimPy system are Resources, Levels and Stores. Each models a congestionpoint where process objects may have to queue up to obtain resources. This section describes the Level type of resourcefacility.

Levels model the production and consumption of a homogeneous undifferentiated “material.” Thus, the currently-available amount of material in a Level resource facility can be fully described by a scalar (real or integer). Processobjects may increase or decrease the currently-available amount of material in a Level facility.

For example, a gasoline station stores gas (petrol) in large tanks. Tankers increase, and refuelled cars decrease, theamount of gas in the station’s storage tanks. Both getting amounts and putting amounts may be subjected to reneginglike requesting amounts from a Resource.

Defining a Level

You define the Level resource facility lev by a statement like this:

lev = Level(name='a_level', unitName='units',capacity='unbounded', initialBuffered=0,putQType=FIFO, getQType=FIFO,monitored=False, monitorType=Monitor)

where

• name (string type) is a descriptive name for the Level object lev is known (e.g., 'inventory').

• unitName (string type) is a descriptive name for the units in which the amount of material in lev is measured(e.g., 'kilograms').

• capacity (positive real or integer) is the capacity of the Level object lev. The default value is set to'unbounded' which is interpreted as sys.maxint.

• initialBuffered (positive real or integer) is the initial amount of material in the Level object lev.

• putQType (FIFO or PriorityQ) is the (producer) queue discipline.

• getQType (FIFO or PriorityQ) is the (consumer) queue discipline.

• monitored (boolean) specifies whether the queues and the amount of material in lev will be recorded.

• monitorType (Monitor or Tally) specifies which type of Recorder to use. Defaults to Monitor.

Every Level resource object, such as lev, also has the following additional attributes:

• lev.amount is the amount currently held in lev.

• lev.putQ is the queue of processes waiting to add amounts to lev, so len(lev.putQ) is the number ofprocesses waiting to add amounts.

• lev.getQ is the queue of processes waiting to get amounts from lev, so len(lev.getQ) is the number ofprocesses waiting to get amounts.

• lev.monitored is True if the queues are to be recorded. In this case lev.putQMon, lev.getQMon,and lev.bufferMon exist.



• lev.putQMon is a Recorder observing lev.putQ.

• lev.getQMon is a Recorder observing lev.getQ.

• lev.bufferMon is a Recorder observing lev.amount.

Getting amounts from a Level

Processes can request amounts from a Level and the same or other processes can offer amounts to it.

A process, the requester, can request an amount ask from the Level resource object lev by a yield get statement.:

• yield get,self,lev,ask[,P]

Here ask must be a positive real or integer (the amount) and P is an optional priority value (real or integer). If lev doesnot hold enough to satisfy the request (that is, ask > lev.amount) the requesting process is passivated and queued (inlev.getQ) in order of its priority. Subject to the priority order, it will be reactivated when there is enough to satisfythe request.

self.got holds the amount actually received by the requester.

..index:: Level; put

Putting amounts into a Level

A process, the offerer, which is usually but not necessarily different from the requester, can offer an amount give to aLevel, lev, by a yield put statement:

• yield put, self,lev,give[,P]

Here give must be a positive real or integer, and P is an optional priority value (real or integer). If the amount offeredwould lead to an overflow (that is, lev.amount + give > lev.capacity) the offering process is passivated and queued(in lev.putQ). Subject to the priority order, it will be reactivated when there is enough space to hold the amountoffered.

The orderings of processes in a Level’s getQ and putQ behave like those described for the waitQ under Resources,except that they are not preemptable. Thus, priority values are ignored when the queue type is FIFO. Otherwise higherpriority values have higher priority, etc.

Example. Suppose that a random demand on an inventory is made each day. Each requested amount is distributednormally with a mean of 1.2 units and a standard deviation of 0.2 units. The inventory (modelled as an object of theLevel class) is refilled by 10 units at fixed intervals of 10 days. There are no back-orders, but a accumulated sum of thetotal stock-out quantities is to be maintained. A trace is to be printed out each day and whenever there is a stock-out:

from random import normalvariate, seedfrom SimPy.Simulation import (Level, Process, activate, get, initialize, hold,

now, put, simulate)

class Deliver(Process):def deliver(self): # an "offeror" PEM

while True:lead = 10.0 # time between refillsdelivery = 10.0 # amount in each refillyield put, self, stock, deliveryprint('at %6.4f, add %6.4f units, now amount = %6.4f' %

(now(), delivery, stock.amount))



yield hold, self, lead

class Demand(Process):stockout = 0.0 # initialize initial stockout amount

def demand(self): # a "requester" PEMday = 1.0 # set time-step to one daywhile True:

yield hold, self, daydd = normalvariate(1.20, 0.20) # today's random demandds = dd - stock.amount# excess of demand over current stock amountif dd > stock.amount: # can't supply requested amount

yield get, self, stock, stock.amount# supply all available amountself.stockout += ds# add unsupplied demand to self.stockoutprint('day %6.4f, demand = %6.4f, shortfall = %6.4f' %

(now(), dd, -ds))else: # can supply requested amount

yield get, self, stock, ddprint('day %6.4f, supplied %6.4f, now amount = %6.4f' %

(now(), dd, stock.amount))

stock = Level(monitored=True) # 'unbounded' capacity and other defaults

seed(99999)initialize()

offeror = Deliver()activate(offeror, offeror.deliver())requester = Demand()activate(requester, requester.demand())


result = (stock.bufferMon.mean(), requester.stockout)print('')print('Summary of results through end of day %6.4f:' % int(now()))print('average stock = %6.4f, cumulative stockout = %6.4f' % result)

Here is the last ten day’s output from one run of this program:

1 at 40.0000, add 10.0000 units, now amount = 10.00002 day 40.0000, supplied 0.7490, now amount = 9.25103 day 41.0000, supplied 1.1651, now amount = 8.08584 day 42.0000, supplied 1.1117, now amount = 6.97415 day 43.0000, supplied 1.1535, now amount = 5.82066 day 44.0000, supplied 0.9202, now amount = 4.90047 day 45.0000, supplied 0.8990, now amount = 4.00148 day 46.0000, supplied 1.1448, now amount = 2.85669 day 47.0000, supplied 1.7287, now amount = 1.1279

10 day 48.0000, supplied 0.9608, now amount = 0.167011 day 49.0000, demand = 0.9837, shortfall = -0.816712

13 Summary of results through end of day 49.0000:



14 average stock = 4.2720, cumulative stockout = 9.7484

[Return to Top ]

Reneging

The yield put can be subject to reneging using one of the compound statements:

• yield (put,self,lev,ask[,P]),(hold,self,waittime)

where if the process does not acquire the amount before waittime is elapsed, the offerer leaves the waitQ and itsexecution continues or

• yield (put,self,lev,ask[,P]),(waitevent,self,events)

where if one of the SimEvents in events occurs before enough becomes available, the offerer leaves the waitQ and itsexecution continues.

In either case if reneging has not occurred the quantity will have been put into the Level and self.stored(lev)will be True. This must be tested immediately after the yield:

1 yield (put, self, lev, ask[,P]), (<reneging clause>)2 if self.stored(lev):3 ## process did not renege4 . . . .5 else:6 ## process reneged before being able to put into the resource

The yield get can also be subject to reneging using one of the compound statements:

• yield (get, self, lev, ask[,P]), (hold, self, waittime)

where if the process does not acquire the amount before waittime is elapsed, the offerer leaves the waitQ and itsexecution continues.

• yield (get,self,lev,ask[,P]),(waitevent,self,events)

where if one of the SimEvents in events occurs before enough becomes available, reneging occurs, the offerer leavesthe waitQ and its execution continues.

In either case if reneging has not occurred self.got == ask and self.acquired(lev) will be True.self.acquired(lev) must be called immediately after the yield:

1 yield (get, self, lev, ask[,P]), (<reneging clause>)2 if self.acquired(lev):3 ## process did not renege, self.got == ask4 . . . .5 else:6 ## process reneged before being able to put into the resource

This test removes the reneging process from the getQ.

[Return to Top ]



2.1.6 Stores

The three resource facilities provided by the SimPy system are Resources, Levels and Stores. Each models a congestionpoint where process objects may have to queue up to obtain resources. This section describes the Store type of resourcefacility.

Stores model the production and consumption of individual items of any Python type. Process objects can insert orremove specific items from the list of items available in a Store. For example, surgical procedures (treated as processobjects) require specific lists of personnel and equipment that may be treated as the items available in a Store type ofresource facility such as a clinic or hospital. As the items held in a Store may be of any Python type, they may inparticular be process objects, and this can be exploited to facilitate modelling Master/Slave relationships. putting andgetting may also be subjected to reneging.

Defining a Store

The Store object sObj is established by a statement like the following:

1 sObj = Store(name='a_store',2 unitName='units',3 capacity='unbounded',4 initialBuffered=None,5 putQType=FIFO,6 getQType=FIFO,7 monitored=False,8 monitorType=Monitor)

where

• name (string type) is a descriptive name for sObj (e.g., 'Inventory').

• unitName (string type) is a descriptive name for the items in sObj (e.g., 'widgets').

• capacity (positive integer) is the maximum number of individual items that can be held in sObj. The defaultvalue is set to 'unbounded' which is interpreted as sys.maxint.

• initialBuffered (a list of individual items) is sObj’s initial content.

• putQType (FIFO or PriorityQ) is the (producer) queue discipline.

• getQType (FIFO or PriorityQ) is the (consumer) queue discipline.

• monitored (boolean) specifies whether sObj’s queues and contents are to be recorded.

• monitorType (Monitor or Tally) specifies the type of Recorder to be used. Defaults to Monitor.

The Store object sObj also has the following additional attributes:

• sObj.theBuffer is a queue (list) of the individual items in sObj. This list is in FIFO order unless the userstores them in a particular order (see Storing objects in an order , below). It is read-only and not directlychangeable by the user.

• sObj.nrBuffered is the current number of objects in sObj. This is read-only and not directly changeableby the user.

• sObj.putQ is the queue of processes waiting to add items to sObj, so that len(sObj.putQ) is the numberof processes waiting to add items.

• sObj.getQ is the queue of processes waiting to get items from sObj, so that len(sObj.getQ) is thenumber of processes waiting to get items.

• If sObj.monitored is True then the queues are to be recorded. In this case sObj.putQMon, sObj.getQMon, and sObj.bufferMon exist.



• sObj.putQMon is a Recorder observing sObj.putQ.

• sObj.getQMon is a Recorder observing sObj.getQ.

• sObj.bufferMon is a Recorder observing sObj.nrBuffered.

Putting objects into a Store

Processes can request items from a Store and the same or other processes can offer items to it. First look at the simplerof these operations, the yield put.

A process, the offerer, which is usually but not necessarily different from the requester, can offer a list of items to sObjby a yield put statement:

• yield put, self, sObj, give[,P]

Here give is a list of any Python objects. If this statement would lead to an overflow (that is, sObj.nrBuffered+ len(give) > sObj.capacity) the putting process is passivated and queued (in sObj.putQ) until there issufficient room. P is an optional priority value (real or integer).

The ordering of processes in a Store’s putQ and getQ behave like those described for the waitQ under Resources ,except that they are not preemptable. Thus, priority values are ignored when the queue type is FIFO. Otherwise higherpriority values indicate higher priority, etc.

The items in sObj are stored in the form of a queue called sObj.theBuffer, which is in FIFO order unless the userhas arranged to sort them into a particular order (see Storing objects in an order below).

Getting objects from a Store

There are two ways of getting objects from a Store. A process, the requester, can either extract the first n objects fromsObj or a list of items chosen by a filter function.

Getting n items is achieved by the following statement:

• yield get, self, sObj, n [,P]

Here n must be a positive integer and P is an optional priority value (real or integer). If sObj does not currently holdenough objects to satisfy this request (that is, n > sObj.nrBuffered) then the requesting process is passivatedand queued (in sObj.getQ). Subject to the priority ordering, it will be reactivated when the request can be satisfied.

The retrieved objects are returned in the list attribute got of the requesting process.

yield get requests with a numerical parameter are honored in priority/FIFO order. Thus, if there are two processesin the Store’s getQ, with the first requesting two items and the second one, the second process gets the requested itemonly after the first process has been given its two items.

Using the get filter function

The second method is to get a list of items chosen by a filter function, written by the user.

The command, using filter function ffn is as follows:

• yield get, self, sObj, ffn [,P]

The user provides a filter function that has a single list argument and returns a list. The argument represents the bufferof the Store. The function must search through the objects in the buffer and return a sub-list of those that satisfy therequirement.



Example The filter function allweight, shown below, is an example of such a filter. The argument, buff, will beautomatically replaced in the execution of yield get,self,store,allweight by the buffer of the Store. Inthis example the objects in the Store are assumed to have weight attributes. The function allweight selects allthose that have a weight attribute over a value W and returns these as a list. The list appears to the calling process asself.got:

1 def allweight(buff):2 """filter: get all items with .weight >=W from store"""3 result = []4 for i in buff:5 if i.weight >= W:6 result.append(i)7 return result

This might be used as follows:

yield get, self, sObj, allweight [,P]

The retrieved objects are returned in the list attribute got of the requesting process.

Note: ‘‘yield get‘‘ requests with a filter function parameter are not necessarily honored in priority/FIFO order, butrather according to the filter function. An example: There are two processes in the Store’s getQ, with the firstrequesting an item with a weight attribute less than 2 kilograms and the second one requesting one with a weightattribute less than 3 kilograms. If there is an item in the Store’s buffer with a weight attribute between 2 and 3 andnone with an attribute of less than 2, the second get requester gets unblocked before the first one. Effectively, theSimPy run time system runs through all processes in the getQ in sequence and tests their filter functions as long asthere are still items in the Store’s buffer.

Example The following program illustrates the use of a Store to model the production and consumption of “widgets”.The widgets are distinguished by their weight:

from SimPy.Simulation import (Lister, Process, Store, activate, get, hold,initialize, now, put, simulate)

class ProducerD(Process):def __init__(self):


def produce(self): # the ProducerD PEMwhile True:

yield put, self, buf, [Widget(9), Widget(7)]yield hold, self, 10

class ConsumerD(Process):def __init__(self):


def consume(self): # the ConsumerD PEMwhile True:

toGet = 3yield get, self, buf, toGetassert len(self.got) == toGetprint('%s Get widget weights %s' % (now(),



[x.weight for x in self.got]))yield hold, self, 11

class Widget(Lister):def __init__(self, weight=0):

self.weight = weight

widgbuf = []for i in range(10):

widgbuf.append(Widget(5))

initialize()

buf = Store(capacity=11, initialBuffered=widgbuf, monitored=True)

for i in range(3): # define and activate 3 producer objectsp = ProducerD()activate(p, p.produce())

for i in range(3): # define and activate 3 consumer objectsc = ConsumerD()activate(c, c.consume())

simulate(until=50)

print('LenBuffer: %s' % buf.bufferMon) # length of bufferprint('getQ: %s' % buf.getQMon) # length of getQprint('putQ %s' % buf.putQMon) # length of putQ

This program produces the following outputs (some lines may be formatted differently):

0 Get widget weights [5, 5, 5]0 Get widget weights [5, 5, 5]0 Get widget weights [5, 5, 5]11 Get widget weights [5, 9, 7]11 Get widget weights [9, 7, 9]11 Get widget weights [7, 9, 7]22 Get widget weights [9, 7, 9]22 Get widget weights [7, 9, 7]22 Get widget weights [9, 7, 9]33 Get widget weights [7, 9, 7]33 Get widget weights [9, 7, 9]40 Get widget weights [7, 9, 7]44 Get widget weights [9, 7, 9]50 Get widget weights [7, 9, 7]LenBuffer: [[0, 10], [0, 7], [0, 9], [0, 11], [0, 8], [0, 10], [0, 7], [10, 9], [10,→˓11], [11, 8], [11, 10], [11, 7], [11, 4], [20, 6], [20, 8], [21, 10], [22, 7], [22,→˓4], [22, 1], [30, 3], [30, 5], [31, 7], [33, 4], [33, 1], [40, 3], [40, 0], [40, 2],→˓ [41, 4], [44, 1], [50, 3], [50, 0], [50, 2]]getQ: [[0, 0], [33, 1], [40, 0], [44, 1], [50, 0]]putQ [[0, 0], [0, 1], [0, 2], [0, 3], [0, 2], [0, 1], [0, 0], [10, 1], [11, 0]]

[Return to Top ]



Reneging

The yield put can be subject to reneging using one of the compound statements:

• yield (put,self,sObj,give [,P]),(hold,self,waittime)

where if the process cannot put the list of objects in give before waittime is elapsed, the offerer leaves the putQ andits execution continues or

• yield (put,self,sObj,give [,P]),(waitevent,self,events)

where if one of the SimEvents in events occurs before it can put the list of objects in give the offerer leaves the putQand its execution continues.

In either case if reneging has not occurred the list of objects in give will have been put into the Store and self.stored(Sobj) will be True.

The mandatory pattern for a put with reneging is:

1 yield (put, self, sObj, give [,P]),(<reneging clause>)2 if self.stored(sObj):3 ## process did not renege4 . . . .5 else:6 ## process reneged before being able to put into the resource

This is so because self.stored( ) not only tests for reneging, but it also cleanly removes a reneging processfrom the putQ.

The yield get can be subject to similar reneging using one of the compound statements:

• yield (get, self, sObj, n [,P]), (hold, self, waittime)

• yield (get, self, sObj, ffn [,P]), (hold, self, waittime)

where if the process does not acquire the amount before waittime is elapsed, the offerer leaves the waitQ and itsexecution continues.

• yield (get, self, sObj, n [,P]), (waitevent, self, events)

• yield (get, self, sObj, ffn [,P]), (waitevent, self, events)

where if one of the SimEvents in events occurs before enough becomes available, reneging occurs, the offerer leavesthe waitQ and its execution continues.

In either case if reneging has not occurred self.got contains the list of retrieved objects and self.acquired(Sobj) will be True.

The mandatory pattern for a get with reneging is:

1 yield (get, self, lev, sObj, <n or ffn> [,P]),(<reneging clause>)2 if self.acquired(sObj):3 ## process did not renege,4 . . . .5 else:6 ## process reneged before being able to put into the resource

This is so because self.acquired( ) not only tests for reneging, but it also cleanly removes a reneging processfrom the getQ.

[Return to Top ]



Storing objects in an order

The contents of a Store instance are listed in a queue. By default, this list is kept in FIFO order. However, the list canbe kept in a user-defined order. You do this by defining a function for reordering the list and adding it to the Storeinstance for which you want to change the list order. Subsequently, the SimPy system will automatically call thatfunction after any addition (put) to the queue.

Example

1 class Parcel:2 def __init__(self, weight):3 self.weight = weight4

5 lightFirst=Store()6

7 def getLightFirst(self, par):8 """Lighter parcels to front of queue"""9 tmplist = [(x.weight, x) for x in par]

10 tmplist.sort()11 return [x for (key, x) in tmplist]12

13 lightFirst.addSort(getLightFirst)

Now any yield get will get the lightest parcel in lightFirst’s queue.

The par parameter is automatically given the Store’s buffer list as value when the SimPy run time system calls there-ordering function.

<aStore>.addSort(<reorderFunction>) adds a re-order function to <aStore>.

Note that such function only changes the sorting order of the Store instance, NOT of the Store class.

Master/Slave modelling with a Store

The items in a Store can be of any Python type. In particular, they may be SimPy processes. This can be used tomodel a Master/Slave situation – an asymmetrical cooperation between two or more processes, with one process (theMaster) being in charge of the cooperation.

The consumer (Master) requests one or more Slaves to be added to the Store’s contents by the Producer (whichmay be the same process as the Slave). For Master/Slave cooperation, the Slave has to be passivated (by a yieldpassivate or yield waitevent statement) after it is put and reactivated when it is retrieved and finishedwith. As this is NOT done automatically by the Store, the Master has to signal the end of the cooperation. ThisMaster/Slave pattern results in the slave process’ life-cycle having a hole between the slave process arrival and itsdeparture after having been served.

Example Cars arrive randomly at a car wash and add themselves to the waitingCars queue. They wait (passively)for a doneSignal. There are two Carwash washers. These get a car, if one is available, wash it, and then sendthe doneSignal to reactivate it. We elect to model the Carwash as Master and the Cars as slaves.

Four cars are put into the waiting list and these make up the initial set of cars waiting for service. Additional carsare generated randomly by the CarGenerator process. Each car yield puts itself onto the waitingCarsStore and immediately passivates itself by waiting for a doneSignal from a car washer. The car washers cycleround getting the next car on the queue, washing it and then sending a doneSignal to it when it has finished:



from SimPy.Simulation import (Process, SimEvent, Store, activate, get,initialize, hold, now, put, simulate, waitevent)

"""Carwash is master"""

class Carwash(Process):"""Carwash is master"""

def __init__(self, name):Process.__init__(self, name=name)

def lifecycle(self):while True:

yield get, self, waitingCars, 1carBeingWashed = self.got[0]yield hold, self, washtimecarBeingWashed.doneSignal.signal(self.name)

class Car(Process):"""Car is slave"""

def __init__(self, name):Process.__init__(self, name=name)self.doneSignal = SimEvent()

def lifecycle(self):yield put, self, waitingCars, [self]yield waitevent, self, self.doneSignalwhichWash = self.doneSignal.signalparamprint('%s car %s done by %s' % (now(), self.name, whichWash))

class CarGenerator(Process):def generate(self):

i = 0while True:

yield hold, self, 2c = Car('%d' % i)activate(c, c.lifecycle())i += 1

washtime = 5initialize()

# put four cars into the queue of waiting carsfor j in range(1, 5):

c = Car(name='%d' % -j)activate(c, c.lifecycle())

waitingCars = Store(capacity=40)for i in range(2):

cw = Carwash('Carwash %s' % i)activate(cw, cw.lifecycle())

cg = CarGenerator()activate(cg, cg.generate())



simulate(until=30)print('waitingCars %s' % [x.name for x in waitingCars.theBuffer])

The output of this program, running to time 30, is:

5 car -1 done by Carwash 05 car -2 done by Carwash 110 car -3 done by Carwash 010 car -4 done by Carwash 115 car 0 done by Carwash 015 car 1 done by Carwash 120 car 2 done by Carwash 020 car 3 done by Carwash 125 car 4 done by Carwash 025 car 5 done by Carwash 130 car 6 done by Carwash 030 car 7 done by Carwash 1waitingCars ['10', '11', '12', '13', '14']

It is also possible to model this car wash with the cars as Master and the Carwash as Slaves.

[Return to Top ]

2.1.7 Random Number Generation

Simulations usually need random numbers. As SimPy does not supply random number generators of its own, usersneed to import them from some other source. Perhaps the most convenient source is the standard Python random mod-ule. It can generate random variates from the following continuous distributions: uniform, beta, exponential, gamma,normal, log-normal, weibull, and vonMises. It can also generate random variates from some discrete distributions.Consult the module’s documentation for details. (Excellent brief descriptions of these distributions, and many others,can be found in the Wikipedia.)

Python’s random module can be used in two ways: you can import the methods directly or you can import theRandom class and make your own random objects. In the second method, each object gives a different randomnumber sequence, thus providing multiple random streams as in Simscript and ModSim.

Here the first method is described (and minimally at that). A single pseudo-random sequence is used for all calls. Youimport the methods you need from the random module. For example:

from random import seed, random, expovariate, normalvariate

In simulation it is good practice to set the initial seed for the pseudo-random sequence at the start of each run. Thenyou have control over the random numbers used. Replications and comparisons are easier and, together with variancereduction techniques, can provide more accurate estimates. In the following code snippet we set the initial seed to333555. X and Y are pseudo-random variates from the two distributions. Both distributions have the same mean:

1 from random import seed, expovariate, normalvariate2

3 seed(333555)4 X = expovariate(0.1)5 Y = normalvariate(10.0, 1.0)

[Return to Top ]


https://docs.python.org/3/library/random.html


https://en.wikipedia.org/wiki/Main_Page


2.1.8 Recording Simulation Results

The Tally and Monitor class objects enable us to observe a single variable of interest and to return a simple datasummary either during or at the completion of a simulation run.

Both use the observe method to record data on one variable. For example we might use a Monitor object to recordthe waiting times for a sequence of customers and another to record the total number of customers in the shop. In adiscrete-event system the number of customers changes only at arrival or departure events and it is at those events thatthe waiting times and number in the shop must be observed. Monitors and Tallys provide elementary statistics usefuleither alone or as the start of a more sophisticated statistical analysis and have proved invaluable in many simulations.

A few more tools associated with recording results are:

• All Monitors are registered automatically in the global list variable allMonitors and all Tallys in variableallTallies. When a simulation is completed results can easily be tabulated and summarized using theselists.

• The function startCollection() can be called to initialize Monitors and Tallys at a certain simulationtime. This is helpful when a simulation needs a ‘warmup’ period to achieve steady state before measurementsare started.

Defining Tallys and Monitors

The ‘’Tally” class records enough information (such as sums and sums of squares) while the simulation runs to returnsimple data summaries. This has the advantage of speed and low memory use. Tallys can also furnish data for ahistogram. However, they do not preserve a time-series usable in more advanced statistical analysis. When a Tally isdefined it is automatically added to the global list allTallies.

To define a new Tally object:

• m = Tally(name='a_Tally', ylab='y', tlab='t')

• name is a descriptive name for the tally object (default=’a_Tally’ ).

• ylab and tlab are descriptive labels used by the SimPy.SimPlot package when plotting graphs ofthe recorded data. They default to 'y' and 't', respectively. (If a histogram is required the methodsetHistogram must be called before recording starts).

The Monitor class preserves a complete time-series of the observed data values, y, and their associated times, t. Itcalculates the data summaries using these series only when they are needed. It is slower and uses more memory thanTally. In long simulations its memory demands may be a disadvantage. When a Monitor is defined it is automaticallyadded to the global list allMonitors.

To define a new Monitor object:

• m = Monitor(name='a_Monitor', ylab='y', tlab='t')

• name is a descriptive name for the Monitor object (default=’a_Monitor’).

• ylab and tlab are descriptive labels used by the SimPy.SimPlot package when plotting graphs of therecorded data. They default to 'y' and 't', respectively. (A histogram can be requested at any time).

Observing data

Both Tallys and Monitors use the observe method to record data. Here and in the next section, r is either a Tally ora Monitor object:

• r.observe(y [,t]) records the current value of the variable, y and time t (or the current time, now( ), ift is missing). A Monitor retains the two values as a sub-list [t,y]. A Tally uses them to update the accumulatedstatistics.



To assure that time averages are calculated correctly observe should be called immediately after a change inthe variable. For example, if we are using Monitor r to record the number N of jobs in a system, the correctsequence of commands on an arrival is:

N = N + 1 # FIRST, increment the number of jobsr.observe(N) # THEN observe the new value of N using r

The recording of data can be reset to start at any time in the simulation:

• r.reset([t]) resets the observations. The recorded data is re-initialized, and the observation starting timeis set to t, or to the current simulation time, now( ), if t is missing.

Data summaries

The following simple data summaries can be obtained from either Monitors or Tallys at any time during or after thesimulation run:

• r.count( ), the current number of observations. (If r is a Monitor this is the same as len(r)).

• r.total( ), the sum of the y values

• r.mean( ), the simple average of the observed y values, ignoring the times at which they were made. Thisis r.total( )/N where N=r.count( ). (If there are no observations, the message: “SimPy: No obser-vations for mean” is printed). See Recording Resource queue lengths for the difference between the simple ornumerical average and the time-average.

��

�

��

��

��

t3t2t1

m.mean()

t4 t5

y3y2

y1

y4

y5

n=5

0

Fig. 2.1: r.mean is the simple average of the y values observed.

• r.var( ) the sample variance of the observations, ignoring the times at which they were made. If an unbiasedestimate of the population variance is desired, the sample variance should be multiplied by n/(n-1), where n =r.count( ). In either case the standard deviation is, of course, the square-root of the variance (If there are noobservations, the message: “SimPy: No observations for sample variance” is printed).

• r.timeAverage([t]) the time-weighted average of y, calculated from time 0 (or the last time r.reset([t]) was called) to time t (or to the current simulation time, now( ), if t is missing). This isdetermined from the area under the graph shown in the figure, divided by the total time of observation. Foraccurate time-average results y most be piecewise constant and observed just after each change in its value.(If there are no observations, the message “SimPy: No observations for timeAverage” is printed. If no time haselapsed, the message “SimPy: No elapsed time for timeAverage” is printed).

• r.timeVariance([t]) the time-weighted variance of the y values calculated from time 0 (or the last timer.reset([t]) was called) to time t (or to the current simulation time, now(), if t is missing).

• r.__str__( ) is a string that briefly describes the current state of the monitor. This can be used in a printstatement.



��

��

��

��

��

t3t2t1 t4 t5

y1y2

y3y4

y5m.timeaverage()

0 t

Fig. 2.2: r.timeAverage( ) is the time-weighted average of the observed y values. Each y value is weighted bythe time for which it exists. The average is the area under the above curve divided by the total time, t.

Special methods for Monitor

The Monitor variety of Recorder is a sub-class of List and has a few extra methods:

• m[i] holds the observation i as a two-item list, [ti, yi]

• m.yseries( ) is a list of the recorded data values, yi

• m.tseries( ) is a list of the recorded times, ti

Histograms

A Histogram is a derived class of list that counts the observations that fall into a number of specified ranges,called bins. A histogram object can be displayed either by printing it out in text form using printHistogrammethod or using the plotHistogram method in the SimPy.SimPlot package.

• h = Histogram(low=<float>, high=<float>, nbins=<integer>) is a histogram object thatcounts the number of y values in each of its bins, based on the recorded y values.

– low is the nominal lowest value of the histogram (default=0.0)

– high is the nominal highest value of the histogram (default=100.0)

– nbins is the number of bins between low and high into which the histogram is to be divided (de-fault=10). SimPy automatically constructs an additional two bins to count the number of y values underthe low value and the number over the high value. Thus, the total number of bins actually used isnbins + 2. The number of y values in each of these bins is counted and assigned to the appropriate bin.

under over

low high

A histogram

nbins = 5

Fig. 2.3: A Histogram contains the number of observed y values falling into each of its nbins+2 bins.

A Histogram, h, can be printed out in text form using

• h.printHistogram(fmt="%s") prints out a histogram in a standard format.

– fmt is a python string format for the bin range values.



Example Printing a histogram from a Tally:

from SimPy.Simulation import Tallyimport random as r

t = Tally(name="myTally", ylab="wait time (sec)")t.setHistogram(low=0.0, high=1.0, nbins=10)for i in range(100000):

t.observe(y=r.random())print(t.printHistogram(fmt="%6.4f"))

This gives a printed histogram like this:

Histogram for myTally:Number of observations: 100000

wait time (sec) < 0.0000: 0 (cum: 0/ 0.0%)0.0000 <= wait time (sec) < 0.1000: 10135 (cum: 10135/ 10.1%)0.1000 <= wait time (sec) < 0.2000: 9973 (cum: 20108/ 20.1%)0.2000 <= wait time (sec) < 0.3000: 10169 (cum: 30277/ 30.3%)0.3000 <= wait time (sec) < 0.4000: 10020 (cum: 40297/ 40.3%)0.4000 <= wait time (sec) < 0.5000: 10126 (cum: 50423/ 50.4%)0.5000 <= wait time (sec) < 0.6000: 9866 (cum: 60289/ 60.3%)0.6000 <= wait time (sec) < 0.7000: 9910 (cum: 70199/ 70.2%)0.7000 <= wait time (sec) < 0.8000: 9990 (cum: 80189/ 80.2%)0.8000 <= wait time (sec) < 0.9000: 9852 (cum: 90041/ 90.0%)0.9000 <= wait time (sec) < 1.0000: 9959 (cum: 100000/100.0%)1.0000 <= wait time (sec) : 0 (cum: 100000/100.0%)

Although both Tallys and Monitors can return a histogram of the data, they furnish histogram data in different ways.

• The Tally object accumulates the histogram’s bin counts as each value is observed during the simulation run.Since none of the individual values are preserved, the setHistogram method must be called to provide ahistogram object to hold the accumulated bin counts before any values are actually observed.

• The Monitor object stores all its data, so the accumulated bin counts can be computed whenever they are desired.Thus, the histogram need not be set up until it is needed and this can be done after the data has been gathered.

Setting up a Histogram for a Tally object

To establish a histogram for a Tally object, r, we call the setHistogram method with appropriate arguments beforewe observe any data, e.g.,

• r.setHistogram(name = '', low=0.0, high=100.0, nbins=10)

As usual, name is a descriptive title for the histogram (defaults to blank). Then, after observing the data:

• h = r.getHistogram( ) returns a completed histogram using the histogram parameters as set up.

Example In the following example we establish a Tally recorder to observe values of an exponential random variate.It uses a histogram with 30 bins (plus the under- and over-count bins):



from SimPy.Simulation import Tallyfrom random import expovariate

r = Tally('Tally') # define a tally object, rr.setHistogram(name='exponential',

low=0.0, high=20.0, nbins=30) # set before observations

for i in range(1000): # make the observationsy = expovariate(0.1)r.observe(y)

h = r.getHistogram() # return the completed histogramprint(h)

Setting up a Histogram for a Monitor object

For Monitor objects, a histogram can be set up and returned in a single call, e.g.,

• h = r.histogram(low=0.0,high=100.0,nbins=10)

This call is equivalent to the following pair:

• r.setHistogram(name = '', low=0.0, high=100.0, nbins=10)

• h = r.getHistogram( ), which returns the completed histogram.

Example Here we establish a Monitor to observe values of an exponential random variate. It uses a histogram with30 bins (plus the under- and over-count bins):

from SimPy.Simulation import Monitorfrom random import expovariate

m = Monitor() # define the Monitor object, m

for i in range(1000): # make the observationsy = expovariate(0.1)m.observe(y)

# set up and return the completed histogramh = m.histogram(low=0.0, high=20, nbins=30)

[Return to Top ]

2.1.9 Other Links

Several example SimPy models are included with the SimPy code distribution in the file SimPyModels.

Klaus Muller and Tony Vignaux, SimPy: Simulating Systems in Python, O’Reilly ONLamp.com, 2003-Feb-27, http://archive.oreilly.com/pub/a/python/2003/02/27/simpy.html

Norman Matloff, Introduction to the SimPy Discrete-Event Simulation Package, U Cal: Davis, 2003, http://heather.cs.ucdavis.edu/~matloff/simcourse.html


http://archive.oreilly.com/pub/a/python/2003/02/27/simpy.html

http://archive.oreilly.com/pub/a/python/2003/02/27/simpy.html

http://heather.cs.ucdavis.edu/~matloff/simcourse.html



David Mertz, Charming Python: SimPy simplifies complex models, IBM Developer Works, Dec 2002, https://www.ibm.com/developerworks/library/l-simpy/index.html

[Return to Top ]

2.1.10 Acknowledgments

We thank those users who have sent comments to correct or improve this text. These include: F. Benichu, BobHelmbold, M. Matti. We will be grateful for further corrections or suggestions.

2.1.11 Appendices

A0. Changes from the previous version of SimPy

SimPy 2.2b1 differs from version 2.1 in the following ways:

Additions:

Changes:

• The Unit tests have been rewritten

• The directory structure of the release has been simplified


A1. SimPy Error Messages

Advisory messages

These messages are returned by simulate( ), as in message=simulate(until=123).

Upon a normal end of a simulation, simulate( ) returns the message:

• SimPy: Normal exit. This means that no errors have occurred and the simulation has run to the time specifiedby the until parameter.

The following messages, returned by simulate( ), are produced at a premature termination of the simulation butallow continuation of the program.

• SimPy: No more events at time x. All processes were completed prior to the endtime given in simu-late(until=endtime).

• SimPy: No activities scheduled. No activities were scheduled when simulate( ) was called.

Fatal error messages

These messages are generated when SimPy-related fatal exceptions occur. They end the SimPy program. Fatal SimPyerror messages are output to sysout.

• Fatal SimPy error: activating function which is not a generator (contains no ‘yield’). A process tried to(re)activate a function which is not a SimPy process (=Python generator). SimPy processes must contain at leastone yield . . . statement.

• Fatal SimPy error: Simulation not initialized. The SimPy program called simulate( ) before calling initialize().


https://www.ibm.com/developerworks/library/l-simpy/index.html

https://www.ibm.com/developerworks/library/l-simpy/index.html


• SimPy: Attempt to schedule event in the past: A yield hold statement has a negative delay time parameter.

• SimPy: initialBuffered exceeds capacity: Attempt to initialize a Store or Level with more units in the bufferthan its capacity allows.

• SimPy: initialBuffered param of Level negative: x: Attempt to initialize a Level with a negative amount x inthe buffer.

• SimPy: Level: wrong type of initialBuffered (parameter=x): Attempt to initialize a buffer with a non-numerical initial buffer content x.

• SimPy: Level: put parameter not a number: Attempt to add a non-numerical amount to a Level’s buffer.

• SimPy: Level: put parameter not positive number: Attempt to add a negative number to a Level’s amount.

• SimPy: Level: get parameter not positive number: x: Attempt to get a negative amount x from a Level.

• SimPy: Store: initialBuffered not a list: Attempt to initialize a Store with other than a list of items in thebuffer.

• SimPy: Item to put missing in yield put stmt: A yield put was malformed by not having a parameter for theitem(s) to put into the Store.

• SimPy: put parameter is not a list: yield put for a Store must have a parameter which is a list of items to putinto the buffer.

• SimPy: Store: get parameter not positive number: x: A yield get for a Store had a negative value for thenumber to get from the buffer.

• SimPy: Fatal error: illegal command: yield x: A yield statement with an undefined command code (firstparameter) x was executed.

Monitor error messages

• SimPy: No observations for mean. No observations were made by the monitor before attempting to calculatethe mean.

• SimPy: No observations for sample variance. No observations were made by the monitor before attemptingto calculate the sample variance.

• SimPy: No observations for timeAverage, No observations were made by the monitor before attempting tocalculate the time-average.

• SimPy: No elapsed time for timeAverage. No simulation time has elapsed before attempting to calculate thetime-average.

A2. SimPy Process States

From the viewpoint of the model builder a SimPy process, p, can at any time be in one of the following states:

• Active: Waiting for a scheduled event. This state simulates an activity in the model. Simulated time passes inthis state. The process state p.active( ) returns True.

• Passive: Not active or terminated. Awaiting (re-)activation by another process. This state simulates a real worldprocess which has not finished and is waiting for some trigger to continue. Does not change simulation time.p.passive( ) returns True.

• Terminated: The process has executed all its action statements. If referenced, it serves as a data instance.p.terminated( ) returns True



Initially (upon creation of the Process instance), a process returns passive.

In addition, a SimPy process, p, can be in the following (sub)states:

• Interrupted: Active process has been interrupted by another process. It can immediately respond to the inter-rupt. This simulates an interruption of a simulated activity before its scheduled completion time. p.interrupted() returns True.

• Queuing: Active process has requested a busy resource and is waiting (passive) to be reactivated upon resourceavailability. p.queuing(a_resource) returns True.

A3. SimPlot, The SimPy plotting utility

SimPlot provides an easy way to graph the results of simulation runs.

A4. SimGUI, The SimPy Graphical User Interface

SimGUI provides a way for users to interact with a SimPy program, changing its parameters and examining the output.

A5. SimulationTrace, the SimPy tracing utility

SimulationTrace has been developed to give users insight into the dynamics of the execution of SimPy simulationprograms. It can help developers with testing and users with explaining SimPy models to themselves and others (e.g.,for documentation or teaching purposes).

A6. SimulationStep, the SimPy event stepping utility

SimulationStep can assist with debugging models, interacting with them on an event-by-event basis, gettingevent-by-event output from a model (e.g. for plotting purposes), etc.

It caters for:

• running a simulation model, while calling a user-defined procedure after every event,

• running a simulation model one event at a time by repeated calls,

• starting and stopping the event-stepping mode under program control.

A7. SimulationRT, a real-time synchronizing utility

SimulationRT allows synchronizing simulation time and real (wall-clock) time. This capability can be used toimplement, e.g., interactive game applications or to demonstrate a model’s execution in real time.

[Return to Top ]

2.1.12 Glossary

(Note: Terms in italics refer to other special terms. Items in code font are code fragments or specific code names.):

activeQ A Resource object automatically creates and maintains its own activeQ, the queue (list) of process objectsthat are currently using one of the Resource’s units. See Resources. (See also the Glossary entry for waitQ.)

activate Commands a process object to being executing its PEM. See Starting and stopping SimPy process objects.



Backus-Naur Form (BNF) notation This manual occasionally uses a modified Backus-Naur Form notation to ex-hibit command syntax, as in the description of the activate command:

activate(p, p.PEM([args]) [,{at=t|delay=period}] [,prior=False])

In this notation, square brackets [ ] indicate items that are optional, braces { } indicate items of which zero ormore may be present, and a vertical bar | indicates a choice between alternatives (with none of them being apossibility).

cancel Deletes all of a process object’s scheduled future events. See Starting and stopping SimPy process objects.

entity An alternative name for process object.

event A SimEvent object. See Advanced synchronization/scheduling capabilities.

FIFO An attribute of a resource object (i.e., a Resource, Level, or Store) indicating that an associated queue (e.g., theActiveQ, waitQ, getQ, or putQ) is to be kept in FIFO order. (See also the Glossary entries for PriorityQ andqType.)

getQ The queue of processes waiting to take something from a Level or Store resource. See also the Glossary entryfor putQ.

interrupt Requests a “victim” process object to interrupt (i.e., to immediately and prematurely end) its current yieldhold,... command. (Note: A process object cannot interrupt itself.) See Asynchronous interruptions.

Level A particular type of resource facility that models the production and consumption of a homogeneous undiffer-entiated “material.” Process objects can increase or decrease the amount of material in a Level resource facility.See Levels.

Monitor A data recorder that compiles basic statistics as a function of time on variables such as waiting times andqueue lengths. (Note: Monitors can also preserve complete time-series data for post-simulation analyses.) SeeRecording Simulation Results.

monitorType The type of Recorder to be used for recording simulation results. Usually this is either a Monitor or aTally. (See also the Glossary entry for Recorder.)

monitored A (boolean) attribute of a resource object indicating whether to keep a record of its activity. See Recorder.

passivate Halts (“freezes”) a process object’s PEM. The process object becomes “passive”. See Starting and stoppingSimPy Process Objects.

PEM An abbreviation for Process Execution Method, q.v.

preempt To force a process object currently using a resource unit to release it and make it available for use by anotherprocess object. See Preemptive requests for a Resource unit.

preemptable A settable attribute of Resource objects. The Resource object’s units are preemptable ifpreemptable==True, otherwise not. See Preemptive requests for a Resource unit.

priority A non-negative integer or real value controlling the order of process objects in a queue. Higher valuesrepresent higher priority. Higher priority process objects are placed ahead of lower priority ones in the queue.See also the Glossary entry for FIFO.

PriorityQ An attribute of a resource object (i.e., a Resource, Level, or Store) indicating that an associated queue (e.g.,the ActiveQ, waitQ, getQ, or putQ) is to be kept in order of priority. (See also the Glossary entries for FIFO,qType.)

process We usually call both process objects and their classes “processes” (with a small “p”). Thus, “process” mayrefer to a Process class or to a process object, depending on context. To avoid ambiguity or for added emphasiswe often explicitly state whether a class or an object is intended.



Process class A class that inherits from SimPy’s Process class and contains at least one Process Execution Method.Process classes may also contain other methods – in particular they may contain an __init__ method. SeeProcesses.

Process Execution Method A Process class method that contains at least one yield ... statement. See Defininga process.

process object An object created from (i.e., an instance of) a Process class. See Processes.

putQ The queue of processes waiting to add something to a Level or Store resource. See also the Glossary entry forgetQ.

reactivate Reactivates (“unfreezes”) a passivated or a terminated process object’s PEM. The process object becomes“active”. See Starting and stopping SimPy Process Objects.

Recorder A device for recording simulation results. Unless otherwise specified, it usually refers either to a Monitoror a Tally. However, Recorders also include histograms and observers. See Recording Simulation Results forMonitors, Tallys, and the other devices for recording simulation results.

renege To leave a queue before acquiring a resource unit. See Reneging – leaving a queue before acquiring a resource.

resource Same as “resource facility.” A congestion point at which process objects may need to queue for access toresources. The term “resource” (with a small “r”) is used as a generic term for the individual resource facilitiesprovided by SimPy (i.e., Resources, Levels, and Stores).

qType An attribute of resource objects indicating whether an associated queue is to be kept in FIFO or PriorityQorder. See the Glossary entries for waitQ, ActiveQ, putQ,and getQ. See also the treatment of these queues in thesections on the individual resources (i.e., Resources, Levels, and Stores).

Resource A particular type of resource facility that possesses several identical resource units. A process object mayacquire one (and only one) of the Resource’s resource units. See Resources .

Resource unit One of the individual resources associated with a Resource type of resource facility. See Resources.

SimEvent The SimPy class for defining and creating SimEvent objects. Occasionally designates a SimEvent objectwhen context makes that usage clear. See Advanced synchronization/scheduling capabilities.

Store A particular type of resource facility that models the production and consumption of individual items. Processobjects can insert or remove items from the Store’s list of available items. See Stores.

Tally A particular type of Recorder that compiles basic statistics as a function of time on variables such as waitingtimes and queue lengths. (Note: Tallys do not preserve complete time-series data for post-simulation analyses.)See Recording Simulation Results. (See also the Glossary entry for monitorType.)

unit (of a Resource) One of the individual resource capabilities provided by a Resource. See Resources.

waitQ A Resource object automatically creates and maintains its own waitQ, the queue (list) of process objects thathave requested but not yet received one of the Resource’s units. See Resources. (See also the Glossary entry foractiveQ.)

2.2 SimPy Classic’s Object Oriented API

Authors


Release 2.3.3

Python Version 2.7 and later

Date Feb 24, 2018

2.2. SimPy Classic’s Object Oriented API 63



Contents

• SimPy Classic’s Object Oriented API

– Introduction

– Basic SimPy OO API Design

– API changes

2.2.1 Introduction

This document describes the object oriented (OO) programming interface introduced with SimPy 2.0. This is anadd-on to the existing API, an alternative API. There is full backward compatibility:

Motivation

Many simulation languages support a procedural modelling style. Using them, problems are decomposed into proce-dures (functions, subroutines) and either represented by general components, such as queues, or represented in codewith data structures.

There are fundamental problems with using the procedural style of modelling and simulation. Procedures do notcorrespond to real world components. Instead, they correspond to methods and algorithms. Mapping from the real(problem) world to the model and back is difficult and not obvious, particularly for users expert in the problem domain,but not in computer science. Perhaps the greatest limitation of the procedural style is the lack of model extensibility.The only way in this style to change simulation models is through functional extension. One can add structuralfunctionality but not alter any of its basic processes.

Right from its beginning, SimPy, on the other hand, has supported an object oriented approach to simulation mod-elling. In SimPy, models can be implemented as collections of autonomous, cooperating objects. These objects areself-sufficient and independent. The actions on these objects are tied to the objects and their attributes. The object-oriented capabilities of Python strongly support this encapsulation.

Why does this matter for simulation models? It helps with the mapping from real-world objects and their activities tomodelled objects and activities, and back. This not only reduces the complexity of the models, it also makes for easiervalidation of models and interpretation of simulation results in real world terms.

The new API allows different, often more concise, cleaner program patterns. It strongly supports the development oflibraries of model components for specific real world domains. It also supports the re-use and extension of modelswhen model specifications change. In particular larger SimPy programs written with the advanced OO API shouldbe easier to maintain and extend. Users are advised to familiarize themselves with this programming paradigm byreading the models in the SimPyModels folder. Most of them are provided in two implementations, i.e. in the existingand in the OO API. Similarly, the programs in the Bank tutorials are provided with both APIs.

The advanced OO API has been developed very elegantly by Stefan Scherfke and Ontje Lünsdorf, starting from SimPy1.9. Thanks, guys, for this great job!

Readers of this document should be familiar with the basics of SimPy and have read at least “Basic SimPy - ManualFor First Time Users”. They should also know how subclassing is done in Python.

2.2.2 Basic SimPy OO API Design

A class Simulation has been added to module SimPy.Simulation. SimulationTrace,SimulationStep and SimulationRT are subclasses of Simulation. Multiple instances of these classes can



co-exist in a SimPy program.

Backward compatibility

Since SimPy 2.0, the package offers both the existing procedural API and an object-oriented API where sim-ulation capabilities are provided by instantiating Simulation. SimulationTrace, SimulationStep orSimulationRT are subclasses of Simulation.

Each SimulationXX instance has its own event list and therefore its own simulation time. A SimulationXXinstance can effectively be considered as a simulated, isolated parallel world. Any Process, Resource, Store, Level,Monitor, Tally or SimEvent instance belongs to one and only one world (i.e., Simulationxx instance).

The following program shows what this means for API and program structure:

from SimPy.Simulation import (Simulation, Process, Resource, request, hold,release)

"""Object Oriented SimPy API"""

# Model components -------------------------------

class Car(Process):def run(self, res):

yield request, self, resyield hold, self, 10yield release, self, resprint("Time: %s" % self.sim.now())

# Model and Experiment ---------------------------

s = Simulation()s.initialize()r = Resource(capacity=5, sim=s)auto = Car(sim=s)s.activate(auto, auto.run(res=r))s.simulate(until=100)

Using the existing API, the following program is semantically the same and also works under the OO version:

from SimPy.Simulation import (activate, hold, initialize, now, request,release, simulate, Process, Resource)

"""Traditional SimPy API"""


class Car(Process):def run(self, res):

yield request, self, resyield hold, self, 10yield release, self, resprint("Time: %s" % now())

# Model and Experiment ---------------------------

initialize()



r = Resource(capacity=5)auto = Car()activate(auto, auto.run(res=r))simulate(until=100)

This full (backwards) compatibility is achieved by the automatic generation of a SimulationXX instance “behind thescenes”.

Models as SimulationXX subclasses

The advanced OO API can be used to generate model classes which are SimulationXX subclasses. This ties a modeland a SimulationXX instance together beautifully. See the following example:

# CarModel.pyimport SimPy.Simulation as simulation"""Advanced Object Oriented SimPy API"""


class Car(simulation.Process):def park(self):

yield simulation.request, self, self.sim.parkingyield simulation.hold, self, 10yield simulation.release, self, self.sim.parkingprint("%s done at %s" % (self.name, self.sim.now()))

# Model ------------------------------------------

class Model(simulation.Simulation):def __init__(self, name, nrCars, spaces):

simulation.Simulation.__init__(self)self.name = nameself.nrCars = nrCarsself.spaces = spaces

def runModel(self):# Initialize Simulation instanceself.initialize()self.parking = simulation.Resource(name="Parking lot",

unitName="spaces",capacity=self.spaces, sim=self)

for i in range(self.nrCars):auto = Car(name="Car%s" % i, sim=self)self.activate(auto, auto.park())

self.simulate(until=100)

if __name__ == "__main__":

# Experiment ----------------------------------myModel = Model(name="Experiment 1", nrCars=10, spaces=5)myModel.runModel()print(myModel.now())

class Model here is a subclass of Simulation. Every model execution, i.e. call to runModel, reinitializes the



simulation (creates an empty event list and sets the time to 0) (see line 24). runModel can thus be called repeatedlyfor multiple runs of the same experiment setup:

if __name__=="__main__":

## Experiments ---------------------------------

myModel = Model(name="Experiment 1", nrCars=10, spaces=5)for repetition in range(100):

## One Experiment -------------------------------

myModel.runModel()print(myModel.now())

Model extension by subclassing

With the advanced OO API, it is now very easy and clean to extend a model by subclassing. This effectively allowsthe creation of model libraries.

For example, the model in the previous example can be extended to one in which also vans compete for parking spaces.This is done by importing the CarModel module and subclassing Model as follows:

# CarModelExtension.py


import SimPy.Simulation as simulationimport CarModel

class Van(simulation.Process):def park(self):

yield simulation.request, self, self.sim.parkingyield simulation.hold, self, 5yield simulation.release, self, self.sim.parkingprint("%s done at %s" % (self.name, self.sim.now()))

# Model ------------------------------------------

class ModelExtension(CarModel.Model):def __init__(self, name, nrCars, spaces, nrTrucks):

CarModel.Model.__init__(self, name=name, nrCars=nrCars, spaces=spaces)self.nrTrucks = nrTrucks

def runModel(self):self.initialize()self.parking = simulation.Resource(name="Parking lot",

capacity=self.spaces,sim=self)

for i in range(self.nrCars):auto = CarModel.Car(name="Car%s" % i, sim=self)self.activate(auto, auto.park())

for i in range(self.nrTrucks):truck = Van(name="Van%s" % i, sim=self)



self.activate(truck, truck.park())self.simulate(until=100)

# Experiment ----------------------------------

myModel1 = ModelExtension(name="Experiment 2", nrCars=10, spaces=5, nrTrucks=3)myModel1.runModel()

Let’s walk through this:

Lines 9-14: Addition of a Van class with a park PEM.

Line 20: Definition of a subclass ModelExtension which extends class Model.

Lines 22-23: Initialization of the model class (Model) from which ModelExtension is derived. When subclass-ing a class in Python, this is always necessary: Python does not automatically initialize the super-class.

Lines 25-36: Defines a runModel method for ModelExtension which also generates and activates Van objects.

2.2.3 API changes

Module SimPy.Simulation

The only change to the API of module SimPy.Simulation is the addition of class Simulation:

1 Module SimPy.Simulation:2 ################ Unchanged ################3 ## yield-verb constants --------------------4 get5 hold6 passivate7 put8 queueevent9 release

10 request11 waitevent12 waituntil13 ## version constant ------------------------14 version15 ## classes ---------------------------------16 FatalSimerror17 Simerror18 ################ Added ################19 Simulation

Thus, after the import:


class Simulation is available to a program.

Actually,:

from SimPy.Simulation import Simulation

is sufficient and even clearer.



class Simulation

The simulation capabilities of a model are provided by instantiating class Simulation like this:


aSimulation = Simulation()## model code follows

Better OO programming style is actually to define a model class which inherits from Simulation:

import SimPy.Simulation as Simulation

class MyModel(Simulation.Simulation):def run(self):

self.initialize()## model code follows

myMo = MyModel()myMo.run()

The self.initialize() is not really necessary, as the Simulation instance is initialized at generation time.If method run for a model (here myMo ) is executed more than once, e.g. for running a simulation repeatedly, self.initialize() resets the model to an empty event list and simulation time 0.

Methods of class Simulation

class Simulation has these methods:

1 class Simulation:2 ## Methods ----------------------------------3 __init__(self)4 initialize(self)5 now(self)6 stopSimulation(self)7 allEventNotices(self)8 allEventTimes(self)9 activate(self, obj, process, at='undefined', delay='undefined', prior=False)

10 reactivate(self, obj, at='undefined', delay='undefined', prior=False)11 startCollection(self, when=0.0, monitors=None, tallies=None)12 simulate(self, until=0)

The semantics and parameters (except for self) of the methods are identical to those of the non-OO SimPy.Simulation functions of the same name. For example, to get the current simulation time of a Simulation object so,the call is:

tcurrent = so.now()

Module SimPy.SimulationTrace

The only change to the API of module SimPy.SimulationTrace is the addition of class SimulationTrace:

1 Module SimPy.SimulationTrace:2 ################ Unchanged ################3 ## yield-verb constants --------------------



4 get5 hold6 passivate7 put8 queueevent9 release

10 request11 waitevent12 waituntil13 ## version constant ------------------------14 version15 ## classes ---------------------------------16 FatalSimerror17 Simerror18 Trace19 ################ Added ################20 SimulationTrace

class SimulationTrace

The simulation capabilities of a model with tracing are provided by instantiating class SimulationTrace like this:

from SimPy.SimulationTrace import SimulationTrace

aSimulation = SimulationTrace()## model code follows

Again, better OO programming style is actually to define a model class which inherits from Simulation:

from SimPy.SimulationTrace import SimulationTrace

class MyModel(SimulationTrace):def run(self):

self.initialize()# model code follows

myMo = MyModel()myMo.run()

class SimulationTrace is a subclass of Simulation and thus provides the same methods, albeit with tracingadded.

The semantics and parameters of the methods are identical to those of the non-OO SimPy.SimulationTracefunctions of the same name.

Methods and attributes of class SimulationTrace

1 class SimulationTrace:2 ## Methods ----------------------------------3 __init__(self)4 initialize(self)5 now(self)6 stopSimulation(self)7 allEventNotices(self)



8 allEventTimes(self)9 activate(self, obj, process, at='undefined', delay='undefined', prior=False)

10 reactivate(self, obj, at='undefined', delay='undefined', prior=False)11 startCollection(self, when=0.0, monitors=None, tallies=None)12 simulate(self, until=0)13 ## trace attribute ---------------------------14 trace

Attribute trace

An initialization of class SimulationTrace generates an instance of class Trace. This becomes an attributetrace of the SimulationTrace instance.

Trace methods

The semantics and parameters of the Trace methods are identical to those of the non-OO SimPy.SimulationTrace trace instance of the same name.

• trace.start(self)

Example:

s.trace.start()

• trace.stop(self)

• trace.treset(self)

• trace.tchange(self, **kmvar)

• trace.ttext(self,par)

Example calls (snippet):

from SimPy.SimulationTrace import SimulationTraces = SimulationTrace()s.initialize()s.trace.ttext("Here we go")

Again, note that you have to qualify the trace instance (see e.g. the last line of the snippet) with theSimulationTrace instance, here s.

Module SimPy.SimulationRT

class SimulationRT

The simulation capabilities plus real time synchronization are provided by instantiating class SimulationRT.

Methods of class SimulationRT

The SimulationRT subclass adds two methods to those inherited from Simulation.

The semantics and parameters of the methods are identical to those of the non-OO SimPy.SimulationRT func-tions of the same name.



• rtnow

• rtset

Example calls (snippet):

from SimPy.SimulationRT import Process, holdclass Car(Process):

def __init__(self):Process.__init__(self, sim=self.sim)

def run(self):print(self.sim.rtnow())yield hold, self, 10

class SimulationStep

The simulation capabilities plus event stepping are provided by instantiating class SimulationStep.

Methods of class SimulationStep

The SimulationStep subclass adds three methods to those inherited from Simulation.

The semantics and parameters of the methods are identical to those of the non-OO SimPy.SimulationStepfunctions of the same name.

• startStepping

• stopStepping

• simulateStep

Example call (snippet):

from SimPy.SimulationStep import *s = SimulationStep()s.initialize()s.simulateStep(until=100, callback=myCallBack)

Classes with a SimulationXX attribute

All SimPy entity (Process, Resource, Store, Level, SimEvent) and monitoring (Monitor, Tally) classes have time-relatedfunctions. In the OO-API of SimPy, they therefore have a .sim attribute which is a reference to the SimulationXXinstance to which they belong. This association is made by providing that reference as a parameter to the constructorof the class.

Important: All class instances instances must refer to the same SimulationXX instance, i.e., their .sim attributesmust have the same value. That value must be the reference to the SimulationXX instance. Any deviation fromthis will lead to strange mis-functioning of a SimPy script.

The constructor calls (signatures) for the classes in question thus change as follows:



class Process

Process.__init__(self, name = 'a_process', sim = None)

Example 1 (snippet):

class Car(Process):def drive(self):

yield hold, self, 10print("Arrived at", self.sim.now())

aSim = Simulation()aSim.initialize()c=Car(name="Mine", sim=aSim)

Example 2, with an __init__ method (snippet):

class Car(Process):def __init__(self, name):

Process.__init__(self, name=name, sim=self.sim)

aSim = Simulation()aSim.initialize()c=Car(name="Mine", whichSim=aSim)

class Resource

Resource.__init__(self, capacity=1, name='a_resource',unitName='units',qType=FIFO, preemptable=0, monitored=False,monitorType=Monitor, sim=None)

Example (snippet):

aSim = Simulation()aSim.initialize()res = Resource(name="Server", sim=aSim)

classes Store and Level

Store:

Store.__init__(self, name=None, capacity='unbounded', unitName='units',putQType=FIFO, getQType=FIFO,monitored=False, monitorType=Monitor, initialBuffered=None,sim=None)

Level:

Level.__init__(self, name=None, capacity='unbounded', unitName='units',putQType=FIFO, getQType=FIFO,monitored=False, monitorType=Monitor, initialBuffered=None,sim=None)



Example (snippet):

aSim = Simulation()aSim.initialize()buffer = Store(name="Parts", sim=aSim)

class SimEvent

SimEvent.__init__(self, name='a_SimEvent', sim=None)

Example (snippet):

aSim = Simulation()aSim.initialize()evt = SimEvent("Boing!", sim=aSim)

classes Monitor and Tally

Monitor:

Monitor.__init__(self, name='a_Monitor', ylab='y', tlab='t', sim=None)

Tally:

Tally.__init__(self, name='a_Tally', ylab='y', tlab='t', sim=None)

Example (snippet):

aSim = Simulation()aSim.initialize()myMoni = Monitor(name="Counting cars", sim=aSim)

2.3 SimPy Classic Simulation with Tracing

Authors


Release 2.3.3



Date December 2011

Updated January 2018

Contents

• SimPy Classic Simulation with Tracing

– Introduction





– Tracing SimPy programs

– trace.tchange(): Changing the tracing

– treset(): Resetting the trace to default values

– trace.tstart(), trace.tstop(): Enabling/disabling the trace

– trace.ttext(): Annotating the trace

– Nice output of class instances

2.3.1 Introduction

The tracing utility has been developed to give users insight into the dynamics of the execution of SimPy simulationprograms. It can help developers with testing and users with explaining SimPy models to themselves and others (e.g.for documentation or teaching purposes).

2.3.2 Tracing SimPy programs

Tracing any SimPy program is as simple as replacing:


with:

from SimPy.SimulationTrace import *

This will give a complete trace of all the scheduling statements executed during the program’s execution.

An even nicer way is to replace this import by:

if __debug__:from SimPy.SimulationTrace import *

else:from SimPy.Simulation import *

This gives a trace during the development and debugging. If one then executes the program with python -O myprog.py,tracing is switched off, and no run-time overhead is incurred. (__debug__ is a global Python constant which is set toFalse by commandline options -O and -OO.)

For the same reason, any user call to trace methods should be written as:

if __debug__:trace.ttext("This will only show during debugging")

Here is an example (bank02.py from the Bank Tutorial):

import SimPy.SimulationTrace as Simulation # <== changed for tracing# import SimPy.Simulation as Simulation

""" Simulate a single customer """

class Customer(Simulation.Process):""" Customer arrives, looks around and leaves """

2.3. SimPy Classic Simulation with Tracing 75


def __init__(self, name):Simulation.Process.__init__(self)self.name = name

def visit(self, timeInBank=0):print("%7.4f %s: Here I am" % (Simulation.now(), self.name))yield Simulation.hold, self, timeInBankprint("%7.4f %s: I must leave" % (Simulation.now(), self.name))

def model():Simulation.initialize()c1 = Customer(name="Klaus")Simulation.activate(c1, c1.visit(timeInBank=10.0), delay=5.0)c2 = Customer(name="Tony")Simulation.activate(c2, c2.visit(timeInBank=8.0), delay=2.0)c3 = Customer(name="Evelyn")Simulation.activate(c3, c3.visit(timeInBank=20.0), delay=12.0)Simulation.simulate(until=400.0)

model()

This program produces the following output:

0 activate <Klaus> at time: 5.0 prior: False0 activate <Tony> at time: 2.0 prior: False0 activate <Evelyn> at time: 12.0 prior: False2.0000 Tony: Here I am

2.0 hold <Tony> delay: 8.05.0000 Klaus: Here I am

5.0 hold <Klaus> delay: 10.010.0000 Tony: I must leave10.0 <Tony> terminated12.0000 Evelyn: Here I am12.0 hold <Evelyn> delay: 20.015.0000 Klaus: I must leave15.0 <Klaus> terminated32.0000 Evelyn: I must leave32.0 <Evelyn> terminated

Another example:

""" bank09.py: Simulate customers arrivingat random, using a Source requesting servicefrom several clerks but a single queuewith a random servicetime

"""from __future__ import generatorsfrom random import Randomimport SimPy.SimulationTrace as Simulation

class Source(Simulation.Process):""" Source generates customers randomly"""

def __init__(self, seed=333):



Simulation.Process.__init__(self)self.SEED = seed

def generate(self, number, interval):rv = Random(self.SEED)for i in range(number):

c = Customer(name="Customer%02d" % (i,))Simulation.activate(c, c.visit(timeInBank=12.0))t = rv.expovariate(1.0 / interval)yield Simulation.hold, self, t

class Customer(Simulation.Process):""" Customer arrives, is served and leaves """


def visit(self, timeInBank=0):arrive = Simulation.now()print("%7.4f %s: Here I am " % (Simulation.now(), self.name))yield Simulation.request, self, counterwait = Simulation.now() - arriveprint("%7.4f %s: Waited %6.3f" % (Simulation.now(),

self.name, wait))tib = counterRV.expovariate(1.0 / timeInBank)yield Simulation.hold, self, tibyield Simulation.release, self, counterprint("%7.4f %s: Finished" % (Simulation.now(), self.name))

def model(counterseed=3939393):global counter, counterRVcounter = Simulation.Resource(name="Clerk", capacity=2) # LcapacitycounterRV = Random(counterseed)Simulation.initialize()sourceseed = 1133source = Source(seed=sourceseed)Simulation.activate(source, source.generate(5, 10.0), 0.0)Simulation.simulate(until=400.0)

model()

This produces:

1 0 activate <a_process> at time: 0 prior: 02 0 activate <Customer00> at time: 0 prior: 03 0 hold <a_process> delay: 8.731404894584 0.0000 Customer00: Here I am5 0 request <Customer00> <Clerk> priority: default6 . . .waitQ: []7 . . .activeQ: ['Customer00']8 0.0000 Customer00: Waited 0.0009 0 hold <Customer00> delay: 8.90355092634

10 8.73140489458 activate <Customer01> at time: 8.73140489458 prior: 011 8.73140489458 hold <a_process> delay: 8.76709801376



12 8.7314 Customer01: Here I am13 8.73140489458 request <Customer01> <Clerk> priority: default14 . . .waitQ: []15 . . .activeQ: ['Customer00', 'Customer01']16 8.7314 Customer01: Waited 0.00017 8.73140489458 hold <Customer01> delay: 21.667688342518 8.90355092634 release <Customer00> <Clerk>19 . . .waitQ: []20 . . .activeQ: ['Customer01']21 8.9036 Customer00: Finished22 8.90355092634 <Customer00> terminated23 17.4985029083 activate <Customer02> at time: 17.4985029083 prior: 024

25 . . . . .

And here is an example showing the trace output for compound yield statements:

import SimPy.SimulationTrace as Simulation

class Client(Simulation.Process):def __init__(self, name):

Simulation.Process.__init__(self, name)

def getServed(self, tank):yield (Simulation.get, self, tank, 10), (Simulation.hold, self, 1.5)if self.acquired(tank):

print("%s got 10 %s" % (self.name, tank.unitName))else:

print("%s reneged" % self.name)

class Filler(Simulation.Process):def __init__(self, name):


def fill(self, tank):for i in range(3):

yield Simulation.hold, self, 1yield Simulation.put, self, tank, 10

Simulation.initialize()tank = Simulation.Level(name="Tank", unitName="gallons")for i in range(2):

c = Client("Client %s" % i)Simulation.activate(c, c.getServed(tank))

f = Filler("Tanker")Simulation.activate(f, f.fill(tank))Simulation.simulate(until=10)

It produces this output:

0 activate <Client 0> at time: 0 prior: False0 activate <Client 1> at time: 0 prior: False0 activate <Tanker> at time: 0 prior: False0 activate <RENEGE - hold for Client 0> at time: 0 prior: False0 get <Client 0>to get: 10 gallons from <Tank> priority: default



. . .getQ: ['Client 0']

. . .putQ: []

. . .in buffer: 0|| RENEGE COMMAND:|| hold <Client 0> delay: 1.50 activate <RENEGE - hold for Client 1> at time: 0 prior: False0 get <Client 1>to get: 10 gallons from <Tank> priority: default. . .getQ: ['Client 0', 'Client 1']. . .putQ: []. . .in buffer: 0|| RENEGE COMMAND:|| hold <Client 1> delay: 1.50 hold <Tanker> delay: 10 hold <RENEGE - hold for Client 0> delay: 1.50 hold <RENEGE - hold for Client 1> delay: 1.51 put <Tanker> to put: 10 gallons into <Tank> priority: default. . .getQ: ['Client 1']. . .putQ: []. . .in buffer: 01 hold <Tanker> delay: 1Client 0 got 10 gallons1 <Client 0> terminated1.5 reactivate <Client 1> time: 1.5 prior: False1.5 <RENEGE - hold for Client 1> terminatedClient 1 reneged1.5 <Client 1> terminated2 put <Tanker> to put: 10 gallons into <Tank> priority: default. . .getQ: []. . .putQ: []. . .in buffer: 102 hold <Tanker> delay: 13 put <Tanker> to put: 10 gallons into <Tank> priority: default. . .getQ: []. . .putQ: []. . .in buffer: 203 <Tanker> terminated

In this example, the Client entities are requesting 10 gallons from the tank (a Level object). If they can’t get them within1.5 time units, they renege (give up waiting). The renege command parts of the compound statements (hold,self,1.5)areshown in the trace output with a prefix of || to indicate that they are being executed in parallel with the primarycommand part (get,self,tank,10). They are being executed by behind-the-scenes processes (e.g. RENEGE-hold forClient 0).

The trace contains all calls of scheduling statements (yield . . . , activate(), reactivate(), cancel() and also the termi-nation of processes (at completion of all their scheduling statements). For yield request and yield release calls, itprovides also the queue status (waiting customers in waitQ and customers being served in activeQ.

2.3.3 trace.tchange(): Changing the tracing

trace is an instance of the Trace class defined in SimulationTrace.py. This gets automatically initialized upon import-ing SimulationTrace..

The tracing can be changed at runtime by calling trace.tchange() with one or more of the following named parameters:

start:

changes the tracing start time. Default is 0. Example: trace.tchange(start=222.2) to starttracing at simulation time 222.2.



end :

changes the tracing end time. Default is a very large number (hopefully past any simulationendtime you will ever use). Example: trace.tchange(end=33) to stop tracing at time 33.

toTrace:

changes the commands to be traced. Default is [“hold”,”activate”,”cancel”,”reactivate”,”passivate”,”request”,“release”,”interrupt”,”waitevent”,”queueevent”, “signal”,”waituntil”,”put”,”get”,”terminated”].Value must be a list containing one or more of those values in the de-fault. Note: “terminated” causes tracing of all process terminations. Example:trace.tchange(toTrace=[“hold”,”activate”]) traces only the yield hold and activate()statements.

outfile:

redirects the trace out put to a file (default is sys.stdout). Value must be a file object open forwriting. Example: trace.tchange(outfile=open(r”c:\python25\bank02trace.txt”,”w”))

All these parameters can be combined. Example: trace.tchange(start=45.0,toTrace=[“terminated”]) will trace allprocess terminations from time 45.0 till the end of the simulation.

The changes become effective at the time trace.tchange() is called. This implies for example that, if the calltrace.tchange(start=50) is made at time 100, it has no effect before now()==100.

2.3.4 treset(): Resetting the trace to default values

The trace parameters can be reset to their default values by calling trace.treset().

2.3.5 trace.tstart(), trace.tstop(): Enabling/disabling the trace

Calling trace.tstart() enables the tracing, and trace.tstop() disables it. Neither call changes any tracing parameters.

2.3.6 trace.ttext(): Annotating the trace

The event-by-event trace output is already very useful in showing the sequence in which SimPy’s quasi-parallel pro-cesses are executed.

For documentation, publishing or teaching purposes, it is even more useful if the trace output can be intermingledwith output which not only shows the command executed, but also contextual information such as the values of statevariables. If one outputs the reason why a specific scheduling command is executed, the trace can give a naturallanguage description of the simulation scenario.

For such in-line annotation, the trace.ttext(<string>) method is available. It provides a string which is output togetherwith the trace of the next scheduling statement. This string is valid only for the scheduling statement following it.

Example:

import SimPy.SimulationTrace as Simulation

class Bus(Simulation.Process):def __init__(self, name):


def operate(self, repairduration=0):tripleft = 1000



while tripleft > 0:Simulation.trace.ttext("Try to go for %s" % tripleft)yield Simulation.hold, self, tripleftif self.interrupted():

tripleft = self.interruptLeftself.interruptReset()Simulation.trace.ttext("Start repair taking %s time units" %

repairduration)yield Simulation.hold, self, repairduration

else:break # no breakdown, ergo bus arrived

Simulation.trace.ttext("<%s> has arrived" % self.name)

class Breakdown(Simulation.Process):def __init__(self, myBus):

Simulation.Process.__init__(self, name="Breakdown " + myBus.name)self.bus = myBus

def breakBus(self, interval):

while True:Simulation.trace.ttext("Breakdown process waiting for %s" %

interval)yield Simulation.hold, self, intervalif self.bus.terminated():

breakSimulation.trace.ttext("Breakdown of %s" % self.bus.name)self.interrupt(self.bus)

print("\n\n+++test_interrupt")Simulation.initialize()b = Bus("Bus 1")Simulation.trace.ttext("Start %s" % b.name)Simulation.activate(b, b.operate(repairduration=20))br = Breakdown(b)Simulation.trace.ttext("Start the Breakdown process for %s" % b.name)Simulation.activate(br, br.breakBus(200))Simulation.trace.start = 100print(Simulation.simulate(until=4000))

This produces:

1 +++test_interrupt2 0 activate <Bus 1> at time: 0 prior: False3 ---- Start Bus 14 0 activate <Breakdown Bus 1> at time: 0 prior: False5 ---- Start the Breakdown process for Bus 16 200 reactivate <Bus 1> time: 200 prior: False7 200 interrupt by: <Breakdown Bus 1> of: <Bus 1>8 ---- Breakdown of Bus 19 200 hold <Breakdown Bus 1> delay: 200

10 ---- Breakdown process waiting for 20011 200 hold <Bus 1> delay: 2012 ---- Start repair taking 20 time units13 220 hold <Bus 1> delay: 80014 ---- Try to go for 800



15 400 reactivate <Bus 1> time: 400 prior: False16 400 interrupt by: <Breakdown Bus 1> of: <Bus 1>17 ---- Breakdown of Bus 118 400 hold <Breakdown Bus 1> delay: 20019

20 . . . . .

The line starting with “—-” is the comment related to the command traced in the preceding output line.

2.3.7 Nice output of class instances

After the import of SimPy.SimulationTrace, all instances of classes Process and Resource (and all their subclasses)have a nice string representation like so:

1 >>> class Bus(SimulationProcess):2 ... def __init__(self, id):3 ... Simulation.Process.__init__(self, name=id)4 ... self.typ = "Bus"5 ...6 >>> b = Bus("Line 15")7 >>> b8 <Instance of Bus, id 21860960:9 .name=Line 15

10 .typ=Bus11 >12 >>>

This can be handy in statements like trace.ttext(“Status of %s”%b).

2.4 Simulation with Real Time Synchronization

Authors



Release 2.3.3


Python-Version 2.7+

Date December 2011


Contents

• Simulation with Real Time Synchronization

– Acknowledgement

– Synchronizing with wall clock time

– Changing the execution speed during a simulation run






– Limitations

– The SimulationRT API

* Structure

* simulate

* rtset

This manual describes SimulationRT, a SimPy module which supports synchronizing the execution of simulationmodels with real (wallclock) time.

2.4.1 Acknowledgement

SimulationRT is based on an idea by Geoff Jarrad of CSIRO (Australia). He contributed a lot to its development andtesting on Windows and Unix.

The code for the adjustment of the execution speed during the simulation run was contributed by Robert C. Ramsdell.

2.4.2 Synchronizing with wall clock time

SimulationRT allows synchronizing simulation time and real (wallclock) time. This capability can be used to imple-ment e.g. interactive game applications or to demonstrate a model’s execution in real time.

It is identical to Simulation, except for the simulate function which takes an additional parameter controlling real-timeexecution speed.

Here is an example:

""" RealTimeFireworks.py """import SimPy.SimulationRT as SimulationRTfrom random import seed, uniformimport time

# Model components -----------------------------------------------------------class Launcher(SimulationRT.Process):

def launch(self):while True:

print("Launch at %2.4f; wallclock: %2.4f" %(SimulationRT.now(),time.clock() - startTime))

yield SimulationRT.hold, self, uniform(1, maxFlightTime)print("Boom!!! Aaaah!! at %2.4f; wallclock: %2.4f" %

(SimulationRT.now(), time.clock() - startTime))

def model():SimulationRT.initialize()for i in range(nrLaunchers):

lau = Launcher()SimulationRT.activate(lau, lau.launch())

SimulationRT.simulate(real_time=True, rel_speed=1, until=20) # unit sim time = 1 sec clock

# Experiment data -----------------------------------------------------------

2.4. Simulation with Real Time Synchronization 83


nrLaunchers = 2maxFlightTime = 5.0startTime = time.clock()seed(1234567)# Experiment -----------------------------------------------------------------model()

rel_speed is the ratio simulated time/wallclock time. rels_speed=1 sets the synchronization so that 1 simulation timeunit is executed in approximately 1 second of wallclock time. Run under Python 2.6 on a Windows Vista-box (2.3GHz), this output resulted over about 17.5 seconds of wallclock time:

Launch at 0.00; wallclock: 0.00Launch at 0.00; wallclock: 0.00Boom!!! Aaaah!! at 1.94; wallclock: 0.00Launch at 1.94; wallclock: 0.00Boom!!! Aaaah!! at 4.85; wallclock: 0.00Launch at 4.85; wallclock: 0.00Boom!!! Aaaah!! at 5.27; wallclock: 0.00Launch at 5.27; wallclock: 0.00Boom!!! Aaaah!! at 6.76; wallclock: 0.00Launch at 6.76; wallclock: 0.00Boom!!! Aaaah!! at 10.14; wallclock: 0.00Launch at 10.14; wallclock: 0.00Boom!!! Aaaah!! at 10.21; wallclock: 0.00Launch at 10.21; wallclock: 0.00Boom!!! Aaaah!! at 11.43; wallclock: 0.00Launch at 11.43; wallclock: 0.00Boom!!! Aaaah!! at 13.34; wallclock: 0.00Launch at 13.34; wallclock: 0.00Boom!!! Aaaah!! at 14.85; wallclock: 0.00Launch at 14.85; wallclock: 0.00Boom!!! Aaaah!! at 17.48; wallclock: 0.00Launch at 17.48; wallclock: 0.00Boom!!! Aaaah!! at 19.15; wallclock: 0.00Launch at 19.15; wallclock: 0.00Boom!!! Aaaah!! at 19.18; wallclock: 0.00Launch at 19.18; wallclock: 0.00

Clearly, the wallclock time does not deviate significantly from the simulation time.

2.4.3 Changing the execution speed during a simulation run

By calling method rtset with a parameter, the ratio simulated time to wallclock time can be changed during a run.

Here is an example:

"""variableTimeRatio.pyShows the SimulationRT capability to change the ratio simulationtime to wallclock time during the run of a simulation."""import SimPy.SimulationRT as SimulationRT

class Changer(SimulationRT.Process):def change(self, when, rat):

global ratioyield SimulationRT.hold, self, when



SimulationRT.rtset(rat)ratio = rat

class Series(SimulationRT.Process):def tick(self, nrTicks):

oldratio = ratiofor i in range(nrTicks):

tLastSim = SimulationRT.now()tLastWallclock = SimulationRT.wallclock()yield SimulationRT.hold, self, 1diffSim = SimulationRT.now() - tLastSimdiffWall = SimulationRT.wallclock() - tLastWallclockprint("now(): %s, sim. time elapsed: %s, wall clock elapsed: "

"%6.3f, sim/wall time ratio: %6.3f" %(SimulationRT.now(), diffSim, diffWall, diffSim / diffWall))

if not ratio == oldratio:print("At simulation time %s: ratio simulation/wallclock "

"time now changed to %s" % (SimulationRT.now(), ratio))oldratio = ratio

SimulationRT.initialize()ticks = 15s = Series()SimulationRT.activate(s, s.tick(nrTicks=ticks))c = Changer()SimulationRT.activate(c, c.change(5, 5))c = Changer()SimulationRT.activate(c, c.change(10, 10))ratio = 1print("At simulation time %s: set ratio simulation/wallclock time to %s" %

(SimulationRT.now(), ratio))SimulationRT.simulate(until=100, real_time=True, rel_speed=ratio)

The program changes the time ratio twice, at simulation times 5 and 10.

When run on a Windows Vista computer under Python 2.7, this results in this output:

At simulation time 0: set ratio simulation/wallclock time to 1now(): 1, sim. time elapsed: 1, wall clock elapsed: 0.998, sim/wall time ratio: 1.→˓002now(): 2, sim. time elapsed: 1, wall clock elapsed: 0.999, sim/wall time ratio: 1.→˓001now(): 3, sim. time elapsed: 1, wall clock elapsed: 0.999, sim/wall time ratio: 1.→˓001now(): 4, sim. time elapsed: 1, wall clock elapsed: 0.999, sim/wall time ratio: 1.→˓001now(): 5, sim. time elapsed: 1, wall clock elapsed: 0.999, sim/wall time ratio: 1.→˓001At simulation time 5: ratio simulation/wallclock time now changed to 5now(): 6, sim. time elapsed: 1, wall clock elapsed: 0.199, sim/wall time ratio: 5.→˓027now(): 7, sim. time elapsed: 1, wall clock elapsed: 0.199, sim/wall time ratio: 5.→˓025now(): 8, sim. time elapsed: 1, wall clock elapsed: 0.199, sim/wall time ratio: 5.→˓026now(): 9, sim. time elapsed: 1, wall clock elapsed: 0.199, sim/wall time ratio: 5.→˓026

2.4. Simulation with Real Time Synchronization 85


now(): 10, sim. time elapsed: 1, wall clock elapsed: 0.199, sim/wall time ratio: 5.→˓024At simulation time 10: ratio simulation/wallclock time now changed to 10now(): 11, sim. time elapsed: 1, wall clock elapsed: 0.099, sim/wall time ratio: 10.→˓108now(): 12, sim. time elapsed: 1, wall clock elapsed: 0.099, sim/wall time ratio: 10.→˓105now(): 13, sim. time elapsed: 1, wall clock elapsed: 0.099, sim/wall time ratio: 10.→˓102now(): 14, sim. time elapsed: 1, wall clock elapsed: 0.099, sim/wall time ratio: 10.→˓104now(): 15, sim. time elapsed: 1, wall clock elapsed: 0.099, sim/wall time ratio: 10.→˓104

2.4.4 Limitations

This module works much better under Windows than under Unix or Linux, i.e., it gives much closer synchronization.Unfortunately, the handling of time in Python is not platform-independent at all. Here is a quote from the documenta-tion of the time module:

1 "clock()2 On Unix, return the current processor time as a floating point number expressed in

→˓seconds.3 The precision, and in fact the very definition of the meaning of ``processor time'' ,

→˓depends4 on that of the C function of the same name, but in any case, this is the function to

→˓use for5 benchmarking Python or timing algorithms.6

7 On Windows, this function returns wall-clock seconds elapsed since the first call to→˓this

8 function, as a floating point number, based on the Win32 function→˓QueryPerformanceCounter().

9 The resolution is typically better than one microsecond.10 "

Also it is deprecated in 3.3 and up.

2.4.5 The SimulationRT API

Structure

Basically, SimulationStep has the same API as Simulation, but with:

• a change in the definition of simulate, and

• an additional method to change execution speed during a simulation run.

simulate

Executes the simulation model.

Call:

simulate(<optional parameters>)



Mandatory parameters: None.

Optional parameters:

• until=0 : the maximum simulation (end) time (positive floating point number; default: 0)

• real_time=False : flag to switch real time synchronization on or off (boolean; default: False, meaning nosynchronization)

• rel_speed=1 : ratio simulation time over wallclock time; example: rel_speed=200 executes 200 units ofsimulation time in about one second (positive floating point number; default: 1, i.e. 1 sec of simulationtime is executed in about 1 sec of wallclock time)

Return value: Simulation status at exit.

rtset

Changes the ratio simulation time over wall clock time.

Call:

rtset(<new ratio>)

Mandatory parameters: None


• rel_speed=1 : ratio simulation time over wallclock time; example: rel_speed=200 executes 200 units ofsimulation time in about one second (positive floating point number; default: 1, i.e. 1 sec of simulationtime is executed in about 1 sec of wallclock time)

Return value: None

2.5 SimPy Classic Simulation with Event Stepping

Authors



Release 2.3.3


Python Version 2.7+

Date December 2011


Contents

• SimPy Classic Simulation with Event Stepping

– Introduction

– SimulationStep overview

– The SimulationStep API

2.5. SimPy Classic Simulation with Event Stepping 87





This manual describes SimulationStep, a SimPy module which supports stepping through a simulation model eventby event.

2.5.1 Introduction

SimulationStep can assist with debugging models, interacting with them on an event-by-event basis, getting event-by-event output from a model (e.g. for plotting purposes), etc.

SimulationStep is a derivative of the Simulation module. Over and above the capabilities provided by Simulation,SimulationStep supports stepping through a simulation model event by event. It caters for:

• running a simulation model, with calling a user-defined procedure after every event,

• running a simulation model one event at a time by repeated calls,

• starting and stopping the event stepping mode under program control.

2.5.2 SimulationStep overview

Here is a simple program which shows basic event stepping capabilities:

# simstep_stepping1.pyimport SimPy.SimulationStep as SimulationStep # (1)

def callbackTimeTrace(): # (2)"""Prints event times"""print("at time=%s" % SimulationStep.now())

class Man(SimulationStep.Process):def walk(self):

print("got up")yield SimulationStep.hold, self, 1print("got to door")yield SimulationStep.hold, self, 10print("got to mail box")yield SimulationStep.hold, self, 10print("got home again")

# trace event timesSimulationStep.initialize()otto = Man()SimulationStep.activate(otto, otto.walk())SimulationStep.startStepping() # (3)SimulationStep.simulate(callback=callbackTimeTrace, until=100) # (4)

A trivial simulation model, but with event stepping:

1. import the stepping version of Simulation

2. define a procedure which gets called after every event

3. switch into event stepping mode

4. run the model with event callback to the procedure defined at (2); simulate in SimulationStep has an extranamed parameter, callback.



Running it produces this output:

got upat time=0got to doorat time=1got to mail boxat time=11got home againat time=21

The callback outputs the simulation time after every event.

Here is another example, the same model, but now with the user getting control back after every event:

# simstep_stepping2.pyimport SimPy.SimulationStep as SimulationStep

def callbackUserControl():"""Allows user to control stepping"""# In python 2.7 you need to make this raw_inputa = input("[Time=%s] Select one: End run (e), Continue stepping (s), "

"Run to end (r)= " % SimulationStep.now())if a == "e":

SimulationStep.stopSimulation()elif a == "s":

returnelse:

SimulationStep.stopStepping()

class Man(SimulationStep.Process):def walk(self):

print("got up")yield SimulationStep.hold, self, 1print("got to door")yield SimulationStep.hold, self, 10print("got to mail box")yield SimulationStep.hold, self, 10print("got home again")

# allow user controlSimulationStep.initialize()otto = Man()SimulationStep.activate(otto, otto.walk())SimulationStep.startStepping()SimulationStep.simulate(callback=callbackUserControl, until=100)

Its interactive output looks like this:

got up[Time=0] Select one: End run (e), Continue stepping (s), Run to end (r)= sgot to door[Time=1] Select one: End run (e), Continue stepping (s), Run to end (r)= sgot to mail box[Time=11] Select one: End run (e), Continue stepping (s), Run to end (r)= s



got home again[Time=21] Select one: End run (e), Continue stepping (s), Run to end (r)= s[Time=21] Select one: End run (e), Continue stepping (s), Run to end (r)= s

or this (the user stopped stepping mode at time=1):

got up[Time=0] Select one: End run (e), Continue stepping (s), Run to end (r)= sgot to door[Time=1] Select one: End run (e), Continue stepping (s), Run to end (r)= rgot to mail boxgot home again

If one wants to run a tested/debugged model full speed, i.e. without stepping, one can write a program as follows:

# simstep_stepping2fast.py

if __debug__:import SimPy.SimulationStep as Simulation

else:import SimPy.Simulation as Simulation

def callbackUserControl():"""Allows user to control stepping"""if __debug__:

# In python 2.7 you need to make this raw_inputa = input("[Time=%s] Select one: End run (e), Continue stepping (s),"

"Run to end (r)= " % Simulation.now())if a == "e":

Simulation.stopSimulation()elif a == "s":

returnelse:

Simulation.stopStepping()

class Man(Simulation.Process):def walk(self):

print("got up")yield Simulation.hold, self, 1print("got to door")yield Simulation.hold, self, 10print("got to mail box")yield Simulation.hold, self, 10print("got home again")

# allow user control if debuggingSimulation.initialize()otto = Man()Simulation.activate(otto, otto.walk())if __debug__:

Simulation.startStepping()Simulation.simulate(callback=callbackUserControl, until=100)

else:Simulation.simulate(until=100)



If one runs this with the Python command line option ‘-O’, any statement starting with if __debug__: is ig-nored/skipped by the Python interpreter.

2.5.3 The SimulationStep API

Structure

Basically, SimulationStep has the same API as Simulation, but with the following additions and changes:

def startStepping() **new**def stopStepping() **new**def simulate() **changed**def simulateStep() **new**

startStepping

Starts the event-stepping.

Call:

startStepping()


Optional parameters: None

Return value: None.

stopStepping

Stops event-stepping.

Call: stopStepping()



Return value: None

simulate

Runs a simulation with callback to a user-defined function after each event, if stepping is turned on. By default,stepping is switched off.

Call: simulate(callback=<proc>,until=<endtime>)



• until = 0: the simulation time until which the simulation is to run (positive floating point or integer number)

• callback = lambda:None: the function to be called after every event (function reference)

Return value: The simulation status at exit (string)



simulateStep

Runs a simulation for one event, with (optional) callback to a user-defined function after the event, if stepping isturned on. By default, stepping is switched off. Thus, to execute the model to completion, simulateStep must be calledrepeatedly.

Note: it is not yet clear to the developers whether this part of the API offers any advantages or capabilities overand above the *simulate* function. The survival of this function in future versions depends on the feedbackfrom the user community.

Call: simulateStep(callback=<proc>,until=<endtime>)



• until = 0: the simulation time until which the simulation is to run (positive floating point or integer number)

• callback = lambda:None: the function to be called after every event (function reference)

Return value: The tuple (simulation status at exit (string),<resumability flag>). <resumability flag> can have oneof two string values: “resumable” if there are more events to be executed, and “notResumable” if all eventshave been exhausted or an error has occurred. simulateStep should normally only be called if “resumable” isreturned.

2.6 Cheatsheets

If you want to have a list of all SimPy commands, their syntax and parameters for your desktop, print out a copy ofthe SimPy Cheatsheet, or better yet bookmark them in your favorite browser.

The Cheatsheet comes as a PDF or Excel:

• PDF A4 page

• Excel A4 page

2.7 Additional examples

The following examples are included in the SimPy source distribution.

2.7.1 Car

from SimPy.Simulation import Process, activate, hold, initialize, now, simulate

class Car(Process):def __init__(self, name, cc):

Process.__init__(self, name=name)self.cc = cc

def go(self):print('%s %s %s' % (now(), self.name, 'Starting'))yield hold, self, 100.0print('%s %s %s' % (now(), self.name, 'Arrived'))


_static/cheatsheet_2_3.xls


initialize()c1 = Car('Car1', 2000) # a new caractivate(c1, c1.go(), at=6.0) # activate at time 6.0c2 = Car('Car2', 1600) # another new caractivate(c2, c2.go()) # activate at time 0simulate(until=200)print('Current time is %s' % now()) # will print 106.0

Output:

0 Car2 Starting6.0 Car1 Starting100.0 Car2 Arrived106.0 Car1 ArrivedCurrent time is 106.0

2.7.2 CarT

from SimPy.SimulationTrace import (Process, activate, initialize, hold, now,simulate)





Output:

0 activate <Car1> at time: 6.0 prior: False0 activate <Car2> at time: 0 prior: False0 Car2 Starting0 hold <Car2> delay: 100.06.0 Car1 Starting6.0 hold <Car1> delay: 100.0100.0 Car2 Arrived100.0 <Car2> terminated106.0 Car1 Arrived106.0 <Car1> terminatedCurrent time is 106.0

2.7. Additional examples 93


2.7.3 Cars




def go(self):print('%s %s %s' % (now(), self.name, 'Starting'))yield request, self, gasstationprint('%s %s %s' % (now(), self.name, 'Got a pump'))yield hold, self, 100.0yield release, self, gasstationprint('%s %s %s' % (now(), self.name, 'Leaving'))

gasstation = Resource(capacity=2, name='gasStation', unitName='pump')initialize()c1 = Car('Car1', 2000)c2 = Car('Car2', 1600)c3 = Car('Car3', 3000)c4 = Car('Car4', 1600)activate(c1, c1.go(), at=4.0) # activate at time 4.0activate(c2, c2.go()) # activate at time 0.0activate(c3, c3.go(), at=3.0) # activate at time 3.0activate(c4, c4.go(), at=3.0) # activate at time 2.0simulate(until=300)print('Current time is %s' % now())

Output:

0 Car2 Starting0 Car2 Got a pump3.0 Car3 Starting3.0 Car3 Got a pump3.0 Car4 Starting4.0 Car1 Starting100.0 Car2 Leaving100.0 Car4 Got a pump103.0 Car3 Leaving103.0 Car1 Got a pump200.0 Car4 Leaving203.0 Car1 LeavingCurrent time is 203.0

2.7.4 CarsT

from SimPy.SimulationTrace import (Process, Resource, activate, initialize,hold, now, release, request, simulate)

class Car(Process):



def __init__(self, name, cc):Process.__init__(self, name=name)self.cc = cc

def go(self):print('%s %s %s' % (now(), self.name, 'Starting'))yield request, self, gasstationprint('%s %s %s' % (now(), self.name, 'Got a pump'))yield hold, self, 100.0yield release, self, gasstationprint('%s %s %s' % (now(), self.name, 'Leaving'))

gasstation = Resource(capacity=2, name='gasStation', unitName='pump')initialize()c1 = Car('Car1', 2000)c2 = Car('Car2', 1600)c3 = Car('Car3', 3000)c4 = Car('Car4', 1600)activate(c1, c1.go(), at=4.0) # activate at time 4.0activate(c2, c2.go()) # activate at time 0.0activate(c3, c3.go(), at=3.0) # activate at time 3.0activate(c4, c4.go(), at=3.0) # activate at time 2.0simulate(until=300)print('Current time is %s' % now())

Output:

0 activate <Car1> at time: 4.0 prior: False0 activate <Car2> at time: 0 prior: False0 activate <Car3> at time: 3.0 prior: False0 activate <Car4> at time: 3.0 prior: False0 Car2 Starting0 request <Car2> <gasStation> priority: default. . .waitQ: []. . .activeQ: ['Car2']0 Car2 Got a pump0 hold <Car2> delay: 100.03.0 Car3 Starting3.0 request <Car3> <gasStation> priority: default. . .waitQ: []. . .activeQ: ['Car2', 'Car3']3.0 Car3 Got a pump3.0 hold <Car3> delay: 100.03.0 Car4 Starting3.0 request <Car4> <gasStation> priority: default. . .waitQ: ['Car4']. . .activeQ: ['Car2', 'Car3']4.0 Car1 Starting4.0 request <Car1> <gasStation> priority: default. . .waitQ: ['Car4', 'Car1']. . .activeQ: ['Car2', 'Car3']100.0 reactivate <Car4> time: 100.0 prior: 1100.0 release <Car2> <gasStation>. . .waitQ: ['Car1']. . .activeQ: ['Car3', 'Car4']100.0 Car2 Leaving100.0 <Car2> terminated



100.0 Car4 Got a pump100.0 hold <Car4> delay: 100.0103.0 reactivate <Car1> time: 103.0 prior: 1103.0 release <Car3> <gasStation>. . .waitQ: []. . .activeQ: ['Car4', 'Car1']103.0 Car3 Leaving103.0 <Car3> terminated103.0 Car1 Got a pump103.0 hold <Car1> delay: 100.0200.0 release <Car4> <gasStation>. . .waitQ: []. . .activeQ: ['Car1']200.0 Car4 Leaving200.0 <Car4> terminated203.0 release <Car1> <gasStation>. . .waitQ: []. . .activeQ: []203.0 Car1 Leaving203.0 <Car1> terminatedCurrent time is 203.0

2.7.5 Carwash

from SimPy.Simulation import (Process, SimEvent, Store, activate, get,initialize, hold, now, put, simulate, waitevent)

"""Carwash is master"""

class Carwash(Process):"""Carwash is master"""

def __init__(self, name):Process.__init__(self, name=name)


yield get, self, waitingCars, 1carBeingWashed = self.got[0]yield hold, self, washtimecarBeingWashed.doneSignal.signal(self.name)

class Car(Process):"""Car is slave"""

def __init__(self, name):Process.__init__(self, name=name)self.doneSignal = SimEvent()

def lifecycle(self):yield put, self, waitingCars, [self]yield waitevent, self, self.doneSignalwhichWash = self.doneSignal.signalparamprint('%s car %s done by %s' % (now(), self.name, whichWash))



class CarGenerator(Process):def generate(self):

i = 0while True:

yield hold, self, 2c = Car('%d' % i)activate(c, c.lifecycle())i += 1

washtime = 5initialize()

# put four cars into the queue of waiting carsfor j in range(1, 5):

c = Car(name='%d' % -j)activate(c, c.lifecycle())

waitingCars = Store(capacity=40)for i in range(2):

cw = Carwash('Carwash %s' % i)activate(cw, cw.lifecycle())

cg = CarGenerator()activate(cg, cg.generate())simulate(until=30)print('waitingCars %s' % [x.name for x in waitingCars.theBuffer])

Output:

5 car -1 done by Carwash 05 car -2 done by Carwash 110 car -3 done by Carwash 010 car -4 done by Carwash 115 car 0 done by Carwash 015 car 1 done by Carwash 120 car 2 done by Carwash 020 car 3 done by Carwash 125 car 4 done by Carwash 025 car 5 done by Carwash 130 car 6 done by Carwash 030 car 7 done by Carwash 1waitingCars ['10', '11', '12', '13', '14']

2.7.6 Breakdown

from SimPy.Simulation import (Process, activate, hold, initialize, now,reactivate, simulate)

class Bus(Process):

def operate(self, repairduration, triplength): # PEMtripleft = triplength # time needed to finish tripwhile tripleft > 0:



yield hold, self, tripleft # try to finish the tripif self.interrupted(): # if another breakdown occurs

print('%s at %s' % (self.interruptCause.name, now()))tripleft = self.interruptLeft # time to finish the tripself.interruptReset() # end interrupt statereactivate(br, delay=repairduration) # restart breakdown bryield hold, self, repairduration # delay for repairsprint('Bus repaired at %s' % now())

else:break # no more breakdowns, bus finished trip

print('Bus has arrived at %s' % now())

class Breakdown(Process):def __init__(self, myBus):

Process.__init__(self, name='Breakdown ' + myBus.name)self.bus = myBus

def breakBus(self, interval): # process execution methodwhile True:

yield hold, self, interval # breakdown interarrivalsif self.bus.terminated():

breakself.interrupt(self.bus) # breakdown to myBus

initialize()b = Bus('Bus') # create a bus objectactivate(b, b.operate(repairduration=20, triplength=1000))br = Breakdown(b) # create breakdown br to bus bactivate(br, br.breakBus(300))simulate(until=4000)print('SimPy: No more events at time %s' % now())

Output:

Breakdown Bus at 300Bus repaired at 320Breakdown Bus at 620Bus repaired at 640Breakdown Bus at 940Bus repaired at 960Bus has arrived at 1060SimPy: No more events at time 1260

2.7.7 Diffpriority

from SimPy.Simulation import (PriorityQ, Process, Resource, activate,initialize, hold, now, release, request,simulate)

class Client(Process):def __init__(self, name):

Process.__init__(self, name)



def getserved(self, servtime, priority, myServer):print('%s requests 1 unit at t=%s' % (self.name, now()))yield request, self, myServer, priorityyield hold, self, servtimeyield release, self, myServerprint('%s done at t=%s' % (self.name, now()))

initialize()# create the *server* Resource objectserver = Resource(capacity=1, qType=PriorityQ, preemptable=1)# create some Client process objectsc1 = Client(name='c1')c2 = Client(name='c2')activate(c1, c1.getserved(servtime=100, priority=1, myServer=server), at=0)activate(c2, c2.getserved(servtime=100, priority=9, myServer=server), at=50)simulate(until=500)

Output:

c1 requests 1 unit at t=0c2 requests 1 unit at t=50c2 done at t=150c1 done at t=200

2.7.8 Firework


class Firework(Process):

def execute(self):print('%s firework launched' % now())yield hold, self, 10.0 # wait 10.0 time unitsfor i in range(10):

yield hold, self, 1.0print('%s tick' % now())

yield hold, self, 10.0 # wait another 10.0 time unitsprint('%s Boom!!' % now())

initialize()f = Firework() # create a Firework object, and# activate it (with some default parameters)activate(f, f.execute(), at=0.0)simulate(until=100)

Output:

0 firework launched11.0 tick12.0 tick13.0 tick14.0 tick15.0 tick



16.0 tick17.0 tick18.0 tick19.0 tick20.0 tick30.0 Boom!!

2.7.9 Levelinventory

from random import normalvariate, seedfrom SimPy.Simulation import (Level, Process, activate, get, initialize, hold,

now, put, simulate)

class Deliver(Process):def deliver(self): # an "offeror" PEM

while True:lead = 10.0 # time between refillsdelivery = 10.0 # amount in each refillyield put, self, stock, deliveryprint('at %6.4f, add %6.4f units, now amount = %6.4f' %

(now(), delivery, stock.amount))yield hold, self, lead

class Demand(Process):stockout = 0.0 # initialize initial stockout amount

def demand(self): # a "requester" PEMday = 1.0 # set time-step to one daywhile True:

yield hold, self, daydd = normalvariate(1.20, 0.20) # today's random demandds = dd - stock.amount# excess of demand over current stock amountif dd > stock.amount: # can't supply requested amount

yield get, self, stock, stock.amount# supply all available amountself.stockout += ds# add unsupplied demand to self.stockoutprint('day %6.4f, demand = %6.4f, shortfall = %6.4f' %

(now(), dd, -ds))else: # can supply requested amount

yield get, self, stock, ddprint('day %6.4f, supplied %6.4f, now amount = %6.4f' %

(now(), dd, stock.amount))

stock = Level(monitored=True) # 'unbounded' capacity and other defaults

seed(99999)initialize()

offeror = Deliver()activate(offeror, offeror.deliver())requester = Demand()



activate(requester, requester.demand())


result = (stock.bufferMon.mean(), requester.stockout)print('')print('Summary of results through end of day %6.4f:' % int(now()))print('average stock = %6.4f, cumulative stockout = %6.4f' % result)

Output:

at 0.0000, add 10.0000 units, now amount = 10.0000day 1.0000, supplied 1.0115, now amount = 8.9885day 2.0000, supplied 1.1377, now amount = 7.8508day 3.0000, supplied 0.7743, now amount = 7.0766day 4.0000, supplied 1.4056, now amount = 5.6709day 5.0000, supplied 0.9736, now amount = 4.6973day 6.0000, supplied 1.1061, now amount = 3.5912day 7.0000, supplied 1.4067, now amount = 2.1845day 8.0000, supplied 1.2884, now amount = 0.8961day 9.0000, demand = 1.4426, shortfall = -0.5465at 10.0000, add 10.0000 units, now amount = 10.0000day 10.0000, supplied 1.5850, now amount = 8.4150day 11.0000, supplied 0.7974, now amount = 7.6176day 12.0000, supplied 0.8854, now amount = 6.7323day 13.0000, supplied 1.4893, now amount = 5.2430day 14.0000, supplied 1.0522, now amount = 4.1908day 15.0000, supplied 1.0492, now amount = 3.1416day 16.0000, supplied 1.2013, now amount = 1.9403day 17.0000, supplied 1.4026, now amount = 0.5377day 18.0000, demand = 1.0874, shortfall = -0.5497day 19.0000, demand = 1.3043, shortfall = -1.3043at 20.0000, add 10.0000 units, now amount = 10.0000day 20.0000, supplied 1.2737, now amount = 8.7263day 21.0000, supplied 0.9550, now amount = 7.7713day 22.0000, supplied 1.3646, now amount = 6.4067day 23.0000, supplied 0.8629, now amount = 5.5438day 24.0000, supplied 1.2893, now amount = 4.2545day 25.0000, supplied 1.1897, now amount = 3.0648day 26.0000, supplied 0.9708, now amount = 2.0940day 27.0000, supplied 1.1397, now amount = 0.9543day 28.0000, demand = 1.1286, shortfall = -0.1743day 29.0000, demand = 0.8891, shortfall = -0.8891at 30.0000, add 10.0000 units, now amount = 10.0000day 30.0000, supplied 0.9496, now amount = 9.0504day 31.0000, supplied 1.1515, now amount = 7.8988day 32.0000, supplied 1.3724, now amount = 6.5264day 33.0000, supplied 0.6977, now amount = 5.8287day 34.0000, supplied 1.2291, now amount = 4.5996day 35.0000, supplied 1.3502, now amount = 3.2494day 36.0000, supplied 1.0396, now amount = 2.2098day 37.0000, supplied 1.3442, now amount = 0.8656day 38.0000, demand = 1.0919, shortfall = -0.2263day 39.0000, demand = 1.4077, shortfall = -1.4077at 40.0000, add 10.0000 units, now amount = 10.0000day 40.0000, supplied 0.9053, now amount = 9.0947day 41.0000, supplied 1.2982, now amount = 7.7965day 42.0000, supplied 1.3458, now amount = 6.4508



day 43.0000, supplied 1.2357, now amount = 5.2151day 44.0000, supplied 1.2822, now amount = 3.9329day 45.0000, supplied 1.1809, now amount = 2.7519day 46.0000, supplied 1.0891, now amount = 1.6629day 47.0000, supplied 1.1926, now amount = 0.4702day 48.0000, demand = 1.3927, shortfall = -0.9224day 49.0000, demand = 1.2690, shortfall = -1.2690

Summary of results through end of day 49.0000:average stock = 4.4581, cumulative stockout = 7.2894

2.7.10 Message


class Message(Process):"""A simple Process"""

def __init__(self, i, len):Process.__init__(self, name='Message' + str(i))self.i = iself.len = len

def go(self):print('%s %s %s' % (now(), self.i, 'Starting'))yield hold, self, 100.0print('%s %s %s' % (now(), self.i, 'Arrived'))

initialize()p1 = Message(1, 203) # new messageactivate(p1, p1.go()) # activate itp2 = Message(2, 33)activate(p2, p2.go(), at=6.0)simulate(until=200)print('Current time is %s' % now()) # will print 106.0

Output:

0 1 Starting6.0 2 Starting100.0 1 Arrived106.0 2 ArrivedCurrent time is 106.0

2.7.11 Monitor

from SimPy.Simulation import Monitorfrom random import expovariate

m = Monitor() # define the Monitor object, m

for i in range(1000): # make the observations



y = expovariate(0.1)m.observe(y)

# set up and return the completed histogramh = m.histogram(low=0.0, high=20, nbins=30)

2.7.12 Resource


class Client(Process):inClients = [] # list the clients in order by their requestsoutClients = [] # list the clients in order by completion of service


def getserved(self, servtime, priority, myServer):Client.inClients.append(self.name)print('%s requests 1 unit at t = %s' % (self.name, now()))# request use of a resource unityield request, self, myServer, priorityyield hold, self, servtime# release the resourceyield release, self, myServerprint('%s done at t = %s' % (self.name, now()))Client.outClients.append(self.name)

initialize()

# the next line creates the ``server`` Resource objectserver = Resource(capacity=2) # server defaults to qType==FIFO

# the next lines create some Client process objectsc1, c2 = Client(name='c1'), Client(name='c2')c3, c4 = Client(name='c3'), Client(name='c4')c5, c6 = Client(name='c5'), Client(name='c6')

# in the next lines each client requests# one of the ``server``'s Resource unitsactivate(c1, c1.getserved(servtime=100, priority=1, myServer=server))activate(c2, c2.getserved(servtime=100, priority=2, myServer=server))activate(c3, c3.getserved(servtime=100, priority=3, myServer=server))activate(c4, c4.getserved(servtime=100, priority=4, myServer=server))activate(c5, c5.getserved(servtime=100, priority=5, myServer=server))activate(c6, c6.getserved(servtime=100, priority=6, myServer=server))

simulate(until=500)

print('Request order: %s' % Client.inClients)print('Service order: %s' % Client.outClients)

Output:



c1 requests 1 unit at t = 0c2 requests 1 unit at t = 0c3 requests 1 unit at t = 0c4 requests 1 unit at t = 0c5 requests 1 unit at t = 0c6 requests 1 unit at t = 0c1 done at t = 100c2 done at t = 100c3 done at t = 200c4 done at t = 200c5 done at t = 300c6 done at t = 300Request order: ['c1', 'c2', 'c3', 'c4', 'c5', 'c6']Service order: ['c1', 'c2', 'c3', 'c4', 'c5', 'c6']

2.7.13 Resource monitor

from math import sqrtfrom SimPy.Simulation import (Monitor, Process, Resource, activate, initialize,

hold, now, release, request, simulate)

class Client(Process):inClients = []outClients = []


def getserved(self, servtime, myServer):print('%s requests 1 unit at t = %s' % (self.name, now()))yield request, self, myServeryield hold, self, servtimeyield release, self, myServerprint('%s done at t = %s' % (self.name, now()))

initialize()

server = Resource(capacity=1, monitored=True, monitorType=Monitor)

c1, c2 = Client(name='c1'), Client(name='c2')c3, c4 = Client(name='c3'), Client(name='c4')

activate(c1, c1.getserved(servtime=100, myServer=server))activate(c2, c2.getserved(servtime=100, myServer=server))activate(c3, c3.getserved(servtime=100, myServer=server))activate(c4, c4.getserved(servtime=100, myServer=server))

simulate(until=500)

print('')print('(TimeAverage no. waiting: %s' % server.waitMon.timeAverage())print('(Number) Average no. waiting: %.4f' % server.waitMon.mean())print('(Number) Var of no. waiting: %.4f' % server.waitMon.var())print('(Number) SD of no. waiting: %.4f' % sqrt(server.waitMon.var()))



print('(TimeAverage no. in service: %s' % server.actMon.timeAverage())print('(Number) Average no. in service: %.4f' % server.actMon.mean())print('(Number) Var of no. in service: %.4f' % server.actMon.var())print('(Number) SD of no. in service: %.4f' % sqrt(server.actMon.var()))print('=' * 40)print('Time history for the "server" waitQ:')print('[time, waitQ]')for item in server.waitMon:

print(item)print('=' * 40)print('Time history for the "server" activeQ:')print('[time, activeQ]')for item in server.actMon:

print(item)

Output:

c1 requests 1 unit at t = 0c2 requests 1 unit at t = 0c3 requests 1 unit at t = 0c4 requests 1 unit at t = 0c1 done at t = 100c2 done at t = 200c3 done at t = 300c4 done at t = 400

(TimeAverage no. waiting: 1.5(Number) Average no. waiting: 1.2857(Number) Var of no. waiting: 1.0612(Number) SD of no. waiting: 1.0302(TimeAverage no. in service: 1.0(Number) Average no. in service: 0.4444(Number) Var of no. in service: 0.2469(Number) SD of no. in service: 0.4969========================================Time history for the "server" waitQ:[time, waitQ][0, 0][0, 1][0, 2][0, 3][100, 2][200, 1][300, 0]========================================Time history for the "server" activeQ:[time, activeQ][0, 0][0, 1][100, 0][100, 1][200, 0][200, 1][300, 0][300, 1][400, 0]



2.7.14 Romulans

from SimPy.Simulation import (Process, initialize, activate, simulate,hold, now, waituntil, stopSimulation)

import random

class Player(Process):

def __init__(self, lives=1, name='ImaTarget'):Process.__init__(self, name)self.lives = lives# provide Player objects with a "damage" propertyself.damage = 0

def life(self):self.message = 'Drat! Some %s survived Federation attack!' % \

(target.name)

def killed(): # function testing for "damage > 5"return self.damage > 5

while True:yield waituntil, self, killedself.lives -= 1self.damage = 0if self.lives == 0:

self.message = '%s wiped out by Federation at \time %s!' % (target.name, now())stopSimulation()

class Federation(Process):

def fight(self): # simulate Federation operationsprint('Three %s attempting to escape!' % (target.name))while True:

if random.randint(0, 10) < 2: # check for hit on playertarget.damage += 1 # hit! increment damage to playerif target.damage <= 5: # target survives

print('Ha! %s hit! Damage= %i' %(target.name, target.damage))

else:if (target.lives - 1) == 0:

print('No more %s left!' % (target.name))else:

print('Now only %i %s left!' % (target.lives - 1,target.name))

yield hold, self, 1

initialize()gameOver = 100# create a Player object named "Romulans"target = Player(lives=3, name='Romulans')activate(target, target.life())# create a Federation object



shooter = Federation()activate(shooter, shooter.fight())simulate(until=gameOver)print(target.message)

Example output, varies:

Three Romulans attempting to escape!Ha! Romulans hit! Damage= 1Ha! Romulans hit! Damage= 2Ha! Romulans hit! Damage= 3Ha! Romulans hit! Damage= 4Ha! Romulans hit! Damage= 5Now only 2 Romulans left!Ha! Romulans hit! Damage= 1Ha! Romulans hit! Damage= 2Ha! Romulans hit! Damage= 3Ha! Romulans hit! Damage= 4Ha! Romulans hit! Damage= 5Now only 1 Romulans left!Ha! Romulans hit! Damage= 1Ha! Romulans hit! Damage= 2Drat! Some Romulans survived Federation attack!

2.7.15 Shopping

from SimPy.Simulation import Process, activate, hold, initialize, simulate

class Customer(Process):def buy(self, budget=0):

print('Here I am at the shops %s' % self.name)t = 5.0for i in range(4):

yield hold, self, t# executed 4 times at intervals of t time unitsprint('I just bought something %s' % self.name)budget -= 10.00

print('All I have left is %s I am going home %s' % (budget, self.name))

initialize()

# create a customer named "Evelyn",C = Customer(name='Evelyn')

# and activate her with a budget of 100activate(C, C.buy(budget=100), at=10.0)


Output:

Here I am at the shops EvelynI just bought something EvelynI just bought something Evelyn



I just bought something EvelynI just bought something EvelynAll I have left is 60.0 I am going home Evelyn

2.7.16 Storewidget

from SimPy.Simulation import (Lister, Process, Store, activate, get, hold,initialize, now, put, simulate)

class ProducerD(Process):def __init__(self):


def produce(self): # the ProducerD PEMwhile True:

yield put, self, buf, [Widget(9), Widget(7)]yield hold, self, 10

class ConsumerD(Process):def __init__(self):


def consume(self): # the ConsumerD PEMwhile True:

toGet = 3yield get, self, buf, toGetassert len(self.got) == toGetprint('%s Get widget weights %s' % (now(),

[x.weight for x in self.got]))yield hold, self, 11

class Widget(Lister):def __init__(self, weight=0):


widgbuf = []for i in range(10):

widgbuf.append(Widget(5))

initialize()

buf = Store(capacity=11, initialBuffered=widgbuf, monitored=True)

for i in range(3): # define and activate 3 producer objectsp = ProducerD()activate(p, p.produce())

for i in range(3): # define and activate 3 consumer objectsc = ConsumerD()activate(c, c.consume())

simulate(until=50)



print('LenBuffer: %s' % buf.bufferMon) # length of bufferprint('getQ: %s' % buf.getQMon) # length of getQprint('putQ %s' % buf.putQMon) # length of putQ

Output:

0 Get widget weights [5, 5, 5]0 Get widget weights [5, 5, 5]0 Get widget weights [5, 5, 5]11 Get widget weights [5, 9, 7]11 Get widget weights [9, 7, 9]11 Get widget weights [7, 9, 7]22 Get widget weights [9, 7, 9]22 Get widget weights [7, 9, 7]22 Get widget weights [9, 7, 9]33 Get widget weights [7, 9, 7]33 Get widget weights [9, 7, 9]40 Get widget weights [7, 9, 7]44 Get widget weights [9, 7, 9]50 Get widget weights [7, 9, 7]LenBuffer: [[0, 10], [0, 7], [0, 9], [0, 11], [0, 8], [0, 10], [0, 7], [10, 9], [10,→˓11], [11, 8], [11, 10], [11, 7], [11, 4], [20, 6], [20, 8], [21, 10], [22, 7], [22,→˓4], [22, 1], [30, 3], [30, 5], [31, 7], [33, 4], [33, 1], [40, 3], [40, 0], [40, 2],→˓ [41, 4], [44, 1], [50, 3], [50, 0], [50, 2]]getQ: [[0, 0], [33, 1], [40, 0], [44, 1], [50, 0]]putQ [[0, 0], [0, 1], [0, 2], [0, 3], [0, 2], [0, 1], [0, 0], [10, 1], [11, 0]]

2.7.17 Tally

from SimPy.Simulation import Tallyimport random as r

t = Tally(name="myTally", ylab="wait time (sec)")t.setHistogram(low=0.0, high=1.0, nbins=10)for i in range(100000):

t.observe(y=r.random())print(t.printHistogram(fmt="%6.4f"))

Output:

Histogram for myTally:Number of observations: 100000

wait time (sec) < 0.0000: 0 (cum: 0/ 0.0%)0.0000 <= wait time (sec) < 0.1000: 10135 (cum: 10135/ 10.1%)0.1000 <= wait time (sec) < 0.2000: 9973 (cum: 20108/ 20.1%)0.2000 <= wait time (sec) < 0.3000: 10169 (cum: 30277/ 30.3%)0.3000 <= wait time (sec) < 0.4000: 10020 (cum: 40297/ 40.3%)0.4000 <= wait time (sec) < 0.5000: 10126 (cum: 50423/ 50.4%)0.5000 <= wait time (sec) < 0.6000: 9866 (cum: 60289/ 60.3%)0.6000 <= wait time (sec) < 0.7000: 9910 (cum: 70199/ 70.2%)0.7000 <= wait time (sec) < 0.8000: 9990 (cum: 80189/ 80.2%)0.8000 <= wait time (sec) < 0.9000: 9852 (cum: 90041/ 90.0%)0.9000 <= wait time (sec) < 1.0000: 9959 (cum: 100000/100.0%)1.0000 <= wait time (sec) : 0 (cum: 100000/100.0%)



Second tally example:

from SimPy.Simulation import Tallyfrom random import expovariate

r = Tally('Tally') # define a tally object, rr.setHistogram(name='exponential',

low=0.0, high=20.0, nbins=30) # set before observations

for i in range(1000): # make the observationsy = expovariate(0.1)r.observe(y)

h = r.getHistogram() # return the completed histogramprint(h)

2.8 SimPy Classic - Examples

These are the examples that ship with the SimPy source.

These pages are to index the examples that ship with SimPy and give users another way to find them.

Contents:

2.8.1 LIST OF MODELS using SimPy Classic

SimPy version 2.3


Python-Version 2.7 or later and 3.x except for the GUI ones

These models are examples of SimPy use written by several authors and usually developed for other purposes, such asteaching and consulting. They are in a variety of styles.

Most models are given in two versions, one with the procedural SimPy API and the other (identified by an “_OO”appended to the program name) with the Object Oriented API introduced in SimPy 2.0.

All of these examples come with SimPY in the docs/examples directory.

NOTE: The SimGUI examples do not work for Python 3 as the SimGUI library has not been ported to Python 3.

New Program Structure

M/M/C Queue: MCC.py, MCC_OO.py

M/M/C (multiple server queue model). This demonstrates both the multiple capacity Resource class and the observemethod of the Monitor class. Random arrivals, exponential service-times. (TV)

Jobs arrive at random into a c-server queue with exponential service-time distribution. Simulate to determine theaverage number in the system and the average time jobs spend in the system.




"""MMC.py

An M/M/c queue

Jobs arrive at random into a c-server queue withexponential service-time distribution. Simulate todetermine the average number in the system andthe average time jobs spend in the system.

- c = Number of servers = 3- rate = Arrival rate = 2.0- stime = mean service time = 1.0

"""from SimPy.Simulation import *from random import expovariate, seed

# Model components ------------------------

class Generator(Process):""" generates Jobs at random """

def execute(self, maxNumber, rate, stime):''' generate Jobs at exponential intervals '''for i in range(maxNumber):

L = Job("Job {0} ".format(i))activate(L, L.execute(stime), delay=0)yield hold, self, expovariate(rate)

class Job(Process):''' Jobs request a gatekeeper and hold

it for an exponential time '''

NoInSystem = 0

def execute(self, stime):arrTime = now()self.trace("Hello World")Job.NoInSystem += 1m.observe(Job.NoInSystem)yield request, self, serverself.trace("At last ")t = expovariate(1.0 / stime)msT.observe(t)yield hold, self, tyield release, self, serverJob.NoInSystem -= 1m.observe(Job.NoInSystem)mT.observe(now() - arrTime)self.trace("Geronimo ")

def trace(self, message):FMT = "{0:7.4f} {1:6} {2:10} ({3:2d})"if TRACING:

print(FMT.format(now(), self.name, message, Job.NoInSystem))# Experiment data -------------------------

2.8. SimPy Classic - Examples 111


TRACING = Falsec = 3 # number of servers in M/M/c systemstime = 1.0 # mean service timerate = 2.0 # mean arrival ratemaxNumber = 1000m = Monitor() # monitor for the number of jobsmT = Monitor() # monitor for the time in systemmsT = Monitor() # monitor for the generated service times

seed(333555777) # seed for random numbers

server = Resource(capacity=c, name='Gatekeeper')

# Model/Experiment ------------------------------

initialize()g = Generator('gen')activate(g, g.execute(maxNumber=maxNumber,

rate=rate, stime=stime))m.observe(0) # number in system is 0 at the startsimulate(until=3000.0)

# Analysis/output -------------------------

print('MMC')print("{0:2d} servers, {1:6.4f} arrival rate, "

"{2:6.4f} mean service time".format(c, rate, stime))print("Average number in the system is {0:6.4f}".format(m.timeAverage()))print("Average time in the system is {0:6.4f}".format(mT.mean()))print("Actual average service-time is {0:6.4f}".format(msT.mean()))

MMC3 servers, 2.0000 arrival rate, 1.0000 mean service time

Average number in the system is 2.8786Average time in the system is 1.4350Actual average service-time is 0.9837

"""MMC_OO.py

An M/M/c queue

Jobs arrive at random into a c-server queue withexponential service-time distribution. Simulate todetermine the average number in the system andthe average time jobs spend in the system.

- c = Number of servers = 3- rate = Arrival rate = 2.0- stime = mean service time = 1.0

"""from SimPy.Simulation import *



from random import expovariate, seed


class Generator(Process):""" generates Jobs at random """

def execute(self, maxNumber, rate, stime):''' generate Jobs at exponential intervals '''for i in range(maxNumber):

L = Job("Job {0}".format(i), sim=self.sim)self.sim.activate(L, L.execute(stime), delay=0)yield hold, self, expovariate(rate)

class Job(Process):''' Jobs request a gatekeeper and hold

it for an exponential time '''

NoInSystem = 0

def execute(self, stime):arrTime = self.sim.now()self.trace("Hello World")Job.NoInSystem += 1self.sim.m.observe(Job.NoInSystem)yield request, self, self.sim.serverself.trace("At last ")t = expovariate(1.0 / stime)self.sim.msT.observe(t)yield hold, self, tyield release, self, self.sim.serverJob.NoInSystem -= 1self.sim.m.observe(Job.NoInSystem)self.sim.mT.observe(self.sim.now() - arrTime)self.trace("Geronimo ")

def trace(self, message):FMT = "{0:7.4f} {1:6} {2:10} {(3:2d)}"if TRACING:

print(FMT.format(self.sim.now(), self.name,message, Job.NoInSystem))

# Experiment data -------------------------

TRACING = Falsec = 3 # number of servers in M/M/c systemstime = 1.0 # mean service timerate = 2.0 # mean arrival ratemaxNumber = 1000seed(333555777) # seed for random numbers

# Model -----------------------------------

class MMCmodel(Simulation):



def run(self):self.initialize()self.m = Monitor(sim=self) # monitor for the number of jobsself.mT = Monitor(sim=self) # monitor for the time in systemself.msT = Monitor(sim=self) # monitor for the generated service timesself.server = Resource(capacity=c, name='Gatekeeper', sim=self)g = Generator(name='gen', sim=self)self.activate(g, g.execute(maxNumber=maxNumber,

rate=rate, stime=stime))self.m.observe(0) # number in system is 0 at the startself.simulate(until=3000.0)

# Experiment ------------------------------model = MMCmodel()model.run()

# Analysis/output -------------------------

print('MMC')print("{0:2d} servers, {1:6.4f} arrival rate, "

"{2:6.4f} mean service time".format(c, rate, stime))print("Average number in the system is {0:6.4f}".format(model.m.timeAverage()))print("Average time in the system is {0:6.4f}".format(model.mT.mean()))print("Actual average service-time is {0:6.4f}".format(model.msT.mean()))

MMC3 servers, 2.0000 arrival rate, 1.0000 mean service time

Average number in the system is 2.8786Average time in the system is 1.4350Actual average service-time is 0.9837

BCC: bcc.py, bcc_OO.py

Determine the probability of rejection of random arrivals to a 2-server system with different service-time distributions.No queues allowed, blocked customers are rejected (BCC). Distributions are Erlang, exponential, and hyperexponen-tial. The theoretical probability is also calculated. (TV)

""" bcc.py

Queue with blocked customers clearedJobs (e.g messages) arrive randomly at rate 1.0 per minute at a2-server system. Mean service time is 0.75 minutes and theservice-time distribution is (1) exponential, (2) Erlang-5, or (3)hyperexponential with p=1/8,m1=2.0, and m2=4/7. However noqueue is allowed; a job arriving when all the servers are busy isrejected.

Develop and run a simulation program to estimate the probability ofrejection (which, in steady-state, is the same as p(c)) Measureand compare the probability for each service time distribution.Though you should test the program with a trace, running just a fewjobs, the final runs should be of 10000 jobs without a trace. Stopthe simulation when 10000 jobs have been generated.

"""



from SimPy.Simulation import *from random import seed, Random, expovariate


dist = ""

def bcc(lam, mu, s):""" bcc - blocked customers cleared model

- returns p[i], i = 0,1,..s.- ps = p[s] = prob of blocking- lameff = effective arrival rate = lam*(1-ps)

See Winston 22.11 for Blocked Customers Cleared Model (Erlang B formula)"""rho = lam / mun = range(s + 1)p = [0] * (s + 1)p[0] = 1sump = 1.0for i in n[1:]:

p[i] = (rho / i) * p[i - 1]sump = sump + p[i]

p0 = 1.0 / sumpfor i in n:

p[i] = p[i] * p0p0 = p[0]ps = p[s]lameff = lam * (1 - ps)L = rho * (1 - ps)return {'lambda': lam, 'mu': mu, 's': s,

'p0': p0, 'p[i]': p, 'ps': ps, 'L': L}

def ErlangVariate(mean, K):""" Erlang random variate

mean = meanK = shape parameterg = rv to be used"""sum = 0.0mu = K / meanfor i in range(K):

sum += expovariate(mu)return (sum)

def HyperVariate(p, m1, m2):""" Hyperexponential random variate

p = prob of branch 1m1 = mean of exponential, branch 1m2 = mean of exponential, branch 2g = rv to be used"""



if random() < p:return expovariate(1.0 / m1)

else:return expovariate(1.0 / m2)

def testHyperVariate():""" tests the HyerVariate rv generator"""ERR = 0x = (1.0981, 1.45546, 5.7470156)p = 0.0, 1.0, 0.5g = Random(1113355)for i in range(3):

x1 = HyperVariate(p[i], 1.0, 10.0, g)# print(p[i], x1)assert abs(x1 - x[i]) < 0.001, 'HyperVariate error'

def erlangB(rho, c):""" Erlang's B formula for probabilities in no-queue

Returns p[n] listsee also SPlus and R version in que.q mmcKque.py has bcc."""n = range(c + 1)pn = list(range(c + 1))term = 1pn[0] = 1sum = 1term = 1.0i = 1while i < (c + 1):

term *= rho / ipn[i] = termsum += pn[i]i += 1

for i in n:pn[i] = pn[i] / sum

return(pn)

class JobGen(Process):""" generates a sequence of Jobs"""

def execute(self, JobRate, MaxJob, mu):global NoInService, Busyfor i in range(MaxJob):

j = Job()activate(j, j.execute(i, mu), delay=0.0)t = expovariate(JobRate)MT.tally(t)yield hold, self, t

self.trace("Job generator finished")

def trace(self, message):if JobGenTRACING:



print("{0:8.4f} \t{1}".format(now(), message))

class Job(Process):""" Jobs that are either accepted or rejected"""

def execute(self, i, mu):""" Job execution, only if accepted"""global NoInService, Busy, dist, NoRejectedif NoInService < c:

self.trace("Job %2d accepted b=%1d" % (i, Busy))NoInService += 1if NoInService == c:

Busy = 1try:

BM.accum(Busy, now())except:

"accum error BM=", BM# yield hold,self,Job.g.expovariate(self.mu);# dist= "Exponential"yield hold, self, ErlangVariate(1.0 / mu, 5)dist = "Erlang "# yield hold,self,HyperVariate(1.0/8,m1=2.0,m2=4.0/7,g=Job.g);# dist= "HyperExpon "NoInService -= 1Busy = 0BM.accum(Busy, now())self.trace("Job %2d leaving b=%1d" % (i, Busy))

else:self.trace("Job %2d REJECT b=%1d" % (i, Busy))NoRejected += 1

def trace(self, message):if JobTRACING:


# Experiment data -------------------------c = 2lam = 1.0 # per minutemu = 1.0 / 0.75 # per minutep = 1.0 / 8m1 = 2.0m2 = 4.0 / 7.0K = 5rho = lam / mu

NoRejected = 0NoInService = 0Busy = 0

JobRate = lamJobMax = 10000

JobTRACING = 0JobGenTRACING = 0




seed(111333)BM = Monitor()MT = Monitor()

initialize()jbg = JobGen()activate(jbg, jbg.execute(1.0, JobMax, mu), 0.0)simulate(until=20000.0)

# Analysis/output -------------------------

print('bcc')print("time at the end = {0}".format(now()))print("now = {0}\tstartTime = {1}".format(now(), BM.startTime))print("No Rejected = {0:d}, ratio= {1}".format(

NoRejected, (1.0 * NoRejected) / JobMax))print("Busy proportion ({0}) = {1:8.6f}".format(dist, BM.timeAverage()))print("Erlang pc (th) = {0:8.6f}".format(erlangB(rho, c)[c]))

bcctime at the end = 10085.751051963694now = 10085.751051963694 startTime = 0.0No Rejected = 1374, ratio= 0.1374Busy proportion (Erlang ) = 0.139016Erlang pc (th) = 0.138462

""" bcc_OO.py

Queue with blocked customers clearedJobs (e.g messages) arrive randomly at rate 1.0 per minute at a2-server system. Mean service time is 0.75 minutes and theservice-time distribution is (1) exponential, (2) Erlang-5, or (3)hyperexponential with p=1/8,m1=2.0, and m2=4/7. However noqueue is allowed; a job arriving when all the servers are busy isrejected.

Develop and run a simulation program to estimate the probability ofrejection (which, in steady-state, is the same as p(c)) Measureand compare the probability for each service time distribution.Though you should test the program with a trace, running just a fewjobs, the final runs should be of 10000 jobs without a trace. Stopthe simulation when 10000 jobs have been generated.

"""

from SimPy.Simulation import *from random import seed, Random, expovariate


dist = ""

def bcc(lam, mu, s):""" bcc - blocked customers cleared model



- returns p[i], i = 0,1,..s.- ps = p[s] = prob of blocking- lameff = effective arrival rate = lam*(1-ps)

See Winston 22.11 for Blocked Customers Cleared Model (Erlang B formula)"""rho = lam / mun = range(s + 1)p = [0] * (s + 1)p[0] = 1sump = 1.0for i in n[1:]:

p[i] = (rho / i) * p[i - 1]sump = sump + p[i]

p0 = 1.0 / sumpfor i in n:

p[i] = p[i] * p0p0 = p[0]ps = p[s]lameff = lam * (1 - ps)L = rho * (1 - ps)return {'lambda': lam, 'mu': mu, 's': s,

'p0': p0, 'p[i]': p, 'ps': ps, 'L': L}

def ErlangVariate(mean, K):""" Erlang random variate

mean = meanK = shape parameterg = rv to be used"""sum = 0.0mu = K / meanfor i in range(K):

sum += expovariate(mu)return (sum)

def HyperVariate(p, m1, m2):""" Hyperexponential random variate

p = prob of branch 1m1 = mean of exponential, branch 1m2 = mean of exponential, branch 2g = rv to be used"""if random() < p:

return expovariate(1.0 / m1)else:

return expovariate(1.0 / m2)

def testHyperVariate():""" tests the HyerVariate rv generator"""ERR = 0x = (1.0981, 1.45546, 5.7470156)



p = 0.0, 1.0, 0.5g = Random(1113355)for i in range(3):

x1 = HyperVariate(p[i], 1.0, 10.0, g)# print(p[i], x1)assert abs(x1 - x[i]) < 0.001, 'HyperVariate error'

def erlangB(rho, c):""" Erlang's B formula for probabilities in no-queue

Returns p[n] listsee also SPlus and R version in que.q mmcKque.py has bcc."""n = range(c + 1)pn = list(range(c + 1))term = 1pn[0] = 1sum = 1term = 1.0i = 1while i < (c + 1):

term *= rho / ipn[i] = termsum += pn[i]i += 1

for i in n:pn[i] = pn[i] / sum

return(pn)

class JobGen(Process):""" generates a sequence of Jobs"""

def execute(self, JobRate, MaxJob, mu):global NoInService, Busyfor i in range(MaxJob):

j = Job(sim=self.sim)self.sim.activate(j, j.execute(i, mu), delay=0.0)t = expovariate(JobRate)MT.tally(t)yield hold, self, t

self.trace("Job generator finished")

def trace(self, message):if JobGenTRACING:

print("{0:8.4f} \t{1}".format(self.sim.now(), message))

class Job(Process):""" Jobs that are either accepted or rejected"""

def execute(self, i, mu):""" Job execution, only if accepted"""global NoInService, Busy, dist, NoRejected



if NoInService < c:self.trace("Job %2d accepted b=%1d" % (i, Busy))NoInService += 1if NoInService == c:

Busy = 1try:

BM.accum(Busy, self.sim.now())except:

"accum error BM=", BM# yield hold,self,Job.g.expovariate(self.mu);# dist= "Exponential"yield hold, self, ErlangVariate(1.0 / mu, 5)dist = "Erlang "# yield hold,self,HyperVariate(1.0/8,m1=2.0,m2=4.0/7,g=Job.g);# dist= "HyperExpon "NoInService -= 1Busy = 0BM.accum(Busy, self.sim.now())self.trace("Job %2d leaving b=%1d" % (i, Busy))

else:self.trace("Job %2d REJECT b=%1d" % (i, Busy))NoRejected += 1

def trace(self, message):if JobTRACING:


# Experiment data -------------------------c = 2lam = 1.0 # per minutemu = 1.0 / 0.75 # per minutep = 1.0 / 8m1 = 2.0m2 = 4.0 / 7.0K = 5rho = lam / mu

NoRejected = 0NoInService = 0Busy = 0

JobRate = lamJobMax = 10000

JobTRACING = 0JobGenTRACING = 0


seed(111333)s = Simulation()BM = Monitor(sim=s)MT = Monitor(sim=s)

s.initialize()jbg = JobGen(sim=s)



s.activate(jbg, jbg.execute(1.0, JobMax, mu), 0.0)s.simulate(until=20000.0)

# Analysis/output -------------------------

print('bcc')print("time at the end = {0}".format(s.now()))print("now = {0}\tstartTime = {1}".format(s.now(), BM.startTime))print("No Rejected = {0:d}, ratio= {1}".format(

NoRejected, (1.0 * NoRejected) / JobMax))print("Busy proportion ({0}) = {1:8.6f}".format(dist, BM.timeAverage()))print("Erlang pc (th) = {0:8.6f}".format(erlangB(rho, c)[c]))

bcctime at the end = 10085.751051963694now = 10085.751051963694 startTime = 0.0No Rejected = 1374, ratio= 0.1374Busy proportion (Erlang ) = 0.139016Erlang pc (th) = 0.138462

callCenter.py, callCenter_OO.py

Scenario: A call center runs around the clock. It has a number of agents online with different skills. Calls by clientswith different questions arrive at an expected rate of callrate per minute (expo. distribution). An agent only deals withclients with questions in his competence areas. The number of agents online and their skills remain constant – whenan agent goes offline, he is replaced by one with the same skills. The expected service time tService[i] per questionfollows an exponential distribution. Clients are impatient and renege if they don’t get service within time tImpatience.

The model returns the frequency distribution of client waiting times, the percentage of reneging clients, and the loadon the agents. This model demonstrates the use of yield get with a filter function. (KGM)

"""callCenter.pyModel shows use of get command with a filter function.

Scenario:A call center runs around the clock. It has a number of agents online withdifferent skills/competences.Calls by clients with different questions arrive at an expected rate ofcallrate per minute (expo. distribution). An agent only deals with clients withquestions in his competence areas. The number of agents online and their skillsremain constant --when an agent goes offline, he is replaced by one withe thesame skills.The expected service time tService[i] per questionfollows an exponential distribution.Clients are impatient and renege if they don't get service within timetImpatience.

* Determine the waiting times of clients.

* Determine the percentage renegers

* Determine the percentage load on agents."""from SimPy.Simulation import *import random as r# Model components -----------------------------------------------------------



class Client(Process):def __init__(self, need):

Process.__init__(self)self.need = need

def getServed(self, callCenter):self.done = SimEvent()callsWaiting = callCenter.callsself.served = Falseself.tArrive = now()yield put, self, callsWaiting, [self]yield hold, self, tImpatience# get here either after renege or after interrupt of# renege==successful callif self.interrupted():

# success, got servicecallCenter.renegeMoni.observe(success)# wait for completion of serviceyield waitevent, self, self.done

else:# renegecallCenter.renegeMoni.observe(renege)callsWaiting.theBuffer.remove(self)callCenter.waitMoni.observe(now() - self.tArrive)if callsWaiting.monitored:

callsWaiting.bufferMon.observe(y=len(callsWaiting.theBuffer))

class CallGenerator(Process):def __init__(self, name, center):

Process.__init__(self, name)self.buffer = center.callsself.center = center

def generate(self):while now() <= endTime:

yield hold, self, r.expovariate(callrate)ran = r.random()for aNeed in clientNeeds:

if ran < probNeeds[aNeed]:need = aNeedbreak

c = Client(need=need)activate(c, c.getServed(callCenter=self.center))

class Agent(Process):def __init__(self, name, skills):

Process.__init__(self, name)self.skills = skillsself.busyMon = Monitor(name="Load on {0}".format(self.name))

def work(self, callCtr):incoming = callCtr.calls

def mySkills(buffer):ret = []for client in buffer:



if client.need in self.skills:ret.append(client)break

return retself.started = now()while True:

self.busyMon.observe(idle)yield get, self, incoming, mySkillsself.busyMon.observe(busy)theClient = self.got[0]callCtr.waitMoni.observe(now() - theClient.tArrive)self.interrupt(theClient) # interrupt the timeout renegeyield hold, self, tService[theClient.need]theClient.done.signal()

class Callcenter:def __init__(self, name):

self.calls = Store(name=name, unitName="call", monitored=True)self.waitMoni = Monitor("Caller waiting time")self.agents = []self.renegeMoni = Monitor("Renegers")

renege = 1success = 0busy = 1idle = 0

# Experiment data ------------------------------------------------------------centerName = "SimCityBank"clientNeeds = ["loan", "insurance", "credit card", "other"]aSkills = [["loan"], ["loan", "credit card"],

["insurance"], ["insurance", "other"]]nrAgents = {0: 1, 1: 2, 2: 2, 3: 2} # skill:nr agents of that skillprobNeeds = {"loan": 0.1, "insurance": 0.2, "credit card": 0.5, "other": 1.0}tService = {"loan": 3., "insurance": 4.,

"credit card": 2., "other": 3.} # minutestImpatience = 3 # minutescallrate = 7. / 10 # Callers per minuteendTime = 10 * 24 * 60 # minutes (10 days)r.seed(12345)

# Model ----------------------------------------------------------------------

def model():initialize()callC = Callcenter(name=centerName)for i in nrAgents.keys(): # loop over skills

for j in range(nrAgents[i]): # loop over nr agents of that skilla = Agent(name="Agent type {0}".format(i), skills=aSkills[i])callC.agents.append(a)activate(a, a.work(callCtr=callC))

# buffer=callC.calls)cg = CallGenerator(name="Call generator", center=callC)activate(cg, cg.generate())simulate(until=endTime)



return callC

for tImpatience in (0.5, 1., 2.,):# Experiment --------------------------------------------------------------callCenter = model()# Analysis/output ---------------------------------------------------------print("\ntImpatience={0} minutes".format(tImpatience))print("==================")callCenter.waitMoni.setHistogram(low=0.0, high=float(tImpatience))try:

print(callCenter.waitMoni.printHistogram(fmt="{0:6.1f}"))except:

passrenegers = [1 for x in callCenter.renegeMoni.yseries() if x == renege]print("\nPercentage reneging callers: {0:4.1f}\n".format(

100.0 * sum(renegers) / callCenter.renegeMoni.count()))for agent in callCenter.agents:

print("Load on {0} (skills= {1}): {2:4.1f} percent".format(agent.name, agent.skills, agent.busyMon.timeAverage() * 100))

tImpatience=0.5 minutes==================

Percentage reneging callers: 10.1

Load on Agent type 0 (skills= ['loan']): 13.9 percentLoad on Agent type 1 (skills= ['loan', 'credit card']): 24.1 percentLoad on Agent type 1 (skills= ['loan', 'credit card']): 24.1 percentLoad on Agent type 2 (skills= ['insurance']): 13.1 percentLoad on Agent type 2 (skills= ['insurance']): 13.1 percentLoad on Agent type 3 (skills= ['insurance', 'other']): 44.3 percentLoad on Agent type 3 (skills= ['insurance', 'other']): 44.3 percent






Load on Agent type 0 (skills= ['loan']): 13.8 percentLoad on Agent type 1 (skills= ['loan', 'credit card']): 24.8 percentLoad on Agent type 1 (skills= ['loan', 'credit card']): 24.8 percentLoad on Agent type 2 (skills= ['insurance']): 13.0 percentLoad on Agent type 2 (skills= ['insurance']): 13.0 percent



Load on Agent type 3 (skills= ['insurance', 'other']): 49.5 percentLoad on Agent type 3 (skills= ['insurance', 'other']): 49.5 percent

Object Oriented version.

"""callCenter_OO.pyModel shows use of get command with a filter function.

Scenario:A call center runs around the clock. It has a number of agents online withdifferent skills/competences.Calls by clients with different questions arrive at an expected rate ofcallrate per minute (expo. distribution). An agent only deals with clients withquestions in his competence areas. The number of agents online and their skillsremainconstant --when an agent goes offline, he is replaced by one withe thesame skills.The expected service time tService[i] per questionfollows an exponential distribution.Clients are impatient and renege if they don't get service within timetImpatience.

* Determine the waiting times of clients.

* Determine the percentage renegers

* Determine the percentage load on agents."""from SimPy.Simulation import *import random as r# Model components -----------------------------------------------------------

class Client(Process):def __init__(self, need, sim):

Process.__init__(self, sim=sim)self.need = need

def getServed(self, callCenter):self.done = SimEvent(sim=self.sim)callsWaiting = callCenter.callsself.served = Falseself.tArrive = self.sim.now()yield put, self, callsWaiting, [self]yield hold, self, tImpatience# get here either after renege or after interrupt of# renege==successful callif self.interrupted():

# success, got servicecallCenter.renegeMoni.observe(success)# wait for completion of serviceyield waitevent, self, self.done

else:# renegecallCenter.renegeMoni.observe(renege)callsWaiting.theBuffer.remove(self)callCenter.waitMoni.observe(self.sim.now() - self.tArrive)if callsWaiting.monitored:

callsWaiting.bufferMon.observe(y=len(callsWaiting.theBuffer))



class CallGenerator(Process):def __init__(self, name, center, sim):

Process.__init__(self, name=name, sim=sim)self.buffer = center.callsself.center = center

def generate(self):while self.sim.now() <= endTime:

yield hold, self, r.expovariate(callrate)ran = r.random()for aNeed in clientNeeds:

if ran < probNeeds[aNeed]:need = aNeedbreak

c = Client(need=need, sim=self.sim)self.sim.activate(c, c.getServed(callCenter=self.center))

class Agent(Process):def __init__(self, name, skills, sim):

Process.__init__(self, name=name, sim=sim)self.skills = skillsself.busyMon = Monitor(

name="Load on {0}".format(self.name), sim=self.sim)

def work(self, callCtr):incoming = callCtr.calls

def mySkills(buffer):ret = []for client in buffer:

if client.need in self.skills:ret.append(client)break

return retself.started = self.sim.now()while True:

self.busyMon.observe(idle)yield get, self, incoming, mySkillsself.busyMon.observe(busy)theClient = self.got[0]callCtr.waitMoni.observe(self.sim.now() - theClient.tArrive)self.interrupt(theClient) # interrupt the timeout renegeyield hold, self, tService[theClient.need]theClient.done.signal()

class Callcenter:def __init__(self, name, sim):

self.calls = Store(name=name, unitName="call", monitored=True, sim=sim)self.waitMoni = Monitor("Caller waiting time", sim=sim)self.agents = []self.renegeMoni = Monitor("Renegers", sim=sim)

renege = 1success = 0



busy = 1idle = 0

# Experiment data ------------------------------------------------------------centerName = "SimCityBank"clientNeeds = ["loan", "insurance", "credit card", "other"]aSkills = [["loan"], ["loan", "credit card"],

["insurance"], ["insurance", "other"]]nrAgents = {0: 1, 1: 2, 2: 2, 3: 2} # skill:nr agents of that skillprobNeeds = {"loan": 0.1, "insurance": 0.2, "credit card": 0.5, "other": 1.0}tService = {"loan": 3., "insurance": 4.,

"credit card": 2., "other": 3.} # minutestImpatienceRange = (0.5, 1., 2.,) # minutescallrate = 7. / 10 # Callers per minuteendTime = 10 * 24 * 60 # minutes (10 days)r.seed(12345)

# Model ----------------------------------------------------------------------

class CallCenterModel(Simulation):def run(self):

self.initialize()callC = Callcenter(name=centerName, sim=self)for i in nrAgents.keys(): # loop over skills

for j in range(nrAgents[i]): # loop over nr agents of that skilla = Agent(name="Agent type {0}".format(

i), skills=aSkills[i], sim=self)callC.agents.append(a)self.activate(a, a.work(callCtr=callC))

# buffer=callC.calls)cg = CallGenerator(name="Call generator", center=callC, sim=self)self.activate(cg, cg.generate())self.simulate(until=endTime)return callC

for tImpatience in tImpatienceRange:# Experiment --------------------------------------------------------------callCenter = CallCenterModel().run()# Analysis/output ---------------------------------------------------------print("\ntImpatience={0} minutes".format(tImpatience))print("==================")callCenter.waitMoni.setHistogram(low=0.0, high=float(tImpatience))try:

print(callCenter.waitMoni.printHistogram(fmt="{0:6.1f}"))except:

passrenegers = [1 for x in callCenter.renegeMoni.yseries() if x == renege]print("\nPercentage reneging callers: {0:4.1f}\n".format(

100.0 * sum(renegers) / callCenter.renegeMoni.count()))for agent in callCenter.agents:

print("Load on {0} (skills= {1}): {2:4.1f} percent".format(agent.name, agent.skills, agent.busyMon.timeAverage() * 100))












cellphone.py, cellphone_OO.py

Simulate the operation of a BCC cellphone system. Calls arrive at random to a cellphone hub with a fixed number ofchannels. Service times are assumed exponential. The objective is to determine the statistics of busy periods in theoperation of a BCC cellphone system.

The program simulates the operation for 10 observation periods and measures the mean and variance of the totaltime blocked, and the number of times blocking occurred in each hour. An observational gap occurs between theobservation periods to make each one’s measurement independent. (TV)

""" cellphone.py

Simulate the operation of a BCC cellphone system.

Calls arrive at random to a cellphone hub with a fixed number ofchannels. Service times are assumed exponential. The objectiveis to determine the statistics of busy periods in the operation of aBCC cellphone system.



The required measurements are(1) the total busy time (all channels full) in each 1-hour period and(2) the total number of busy times in a 1-hour period.

The simulation is continuous but the observing Process, aStatistician, breaks the time into 1-hour observation periodsseparated by 15-minute gaps to reduce autocorrelation. The total busytime and number of busy times in each interval is printed.

"""from SimPy.Simulation import *import random as ran


class CallSource(Process):""" generates a sequence of calls """

def execute(self, maxN, lam, cell):for i in range(maxN):

j = Call("Call{0:03d}".format(i))activate(j, j.execute(cell))yield hold, self, ran.expovariate(lam)

class Call(Process):""" Calls arrive at random at the cellphone hub"""

def execute(self, cell):self.trace("arrived")if cell.Nfree == 0:

self.trace("blocked and left")else:

self.trace("got a channel")cell.Nfree -= 1if cell.Nfree == 0:

self.trace("start busy period======")cell.busyStartTime = now()cell.totalBusyVisits += 1

yield hold, self, ran.expovariate(mu)self.trace("finished")if cell.Nfree == 0:

self.trace("end busy period++++++")cell.busyEndTime = now()busy = now() - cell.busyStartTimeself.trace(" busy = {0:9.4f}".format(busy))cell.totalBusyTime += busy

cell.Nfree += 1

def trace(self, message):if TRACING:

print("{0:7.4f} {1:13s} {2} ".format(now(), message, self.name))

class Cell:""" Holds global measurements"""Nfree = 0



totalBusyTime = 0.0totalBusyVisits = 0result = ()

class Statistician(Process):""" observes the system at intervals """

def execute(self, Nperiods, obsPeriod, obsGap, cell):cell.busyEndTime = now() # simulation start timeif STRACING:

print("Busy time Number")for i in range(Nperiods):

yield hold, self, obsGapcell.totalBusyTime = 0.0cell.totalBusyVisits = 0if cell.Nfree == 0:

cell.busyStartTime = now()yield hold, self, obsPeriodif cell.Nfree == 0:

cell.totalBusyTime += now() - cell.busyStartTimeif STRACING:

print("{0:7.3f} {1:5d}".format(cell.totalBusyTime, cell.totalBusyVisits))

m.tally(cell.totalBusyTime)bn.tally(cell.totalBusyVisits)

stopSimulation()cell.result = (m.mean(), m.var(), bn.mean(), bn.var())

# Experiment data -------------------------

NChannels = 4 # number of channels in the cellmaxN = 10000ranSeed = 3333333lam = 1.0 # per minutemu = 0.6667 # per minuteNperiods = 10obsPeriod = 60.0 # minutesobsGap = 15.0 # gap between observation periods

TRACING = FalseSTRACING = True

# Experiment ------------------------------

m = Monitor()bn = Monitor()ran.seed(ranSeed)

cell = Cell() # the cellphone towercell.Nfree = NChannels

initialize()s = Statistician('Statistician')activate(s, s.execute(Nperiods, obsPeriod, obsGap, cell))g = CallSource('CallSource')



activate(g, g.execute(maxN, lam, cell))simulate(until=10000.0)

# Output -------------------------print('cellphone')# input data:print("lambda mu s Nperiods obsPeriod obsGap")FMT = "{0:7.4f} {1:6.4f} {2:4d} {3:4d} {4:6.2f} {5:6.2f}"print(FMT.format(lam, mu, NChannels, Nperiods, obsPeriod, obsGap))

sr = cell.resultprint("Busy Time: mean = {0:6.3f} var= {1:6.3f}".format(sr[0], sr[1]))print("Busy Number: mean = {0:6.3f} var= {1:6.3f}".format(sr[2], sr[3]))

Busy time Number5.857 152.890 61.219 41.512 74.237 51.140 13.596 83.964 84.146 92.091 5

cellphonelambda mu s Nperiods obsPeriod obsGap1.0000 0.6667 4 10 60.00 15.00

Busy Time: mean = 3.065 var= 2.193Busy Number: mean = 6.800 var= 12.360

Object orientated

""" cellphone_OO.py

Simulate the operation of a BCC cellphone system.

Calls arrive at random to a cellphone hub with a fixed number ofchannels. Service times are assumed exponential. The objectiveis to determine the statistics of busy periods in the operation of aBCC cellphone system.

The required measurements are(1) the total busy time (all channels full) in each 1-hour period and(2) the total number of busy times in a 1-hour period.

The simulation is continuous but the observing Process, aStatistician, breaks the time into 1-hour observation periodsseparated by 15-minute gaps to reduce autocorrelation. The total busytime and number of busy times in each interval is printed.





class CallSource(Process):""" generates a sequence of calls """

def execute(self, maxN, lam, cell):for i in range(maxN):

j = Call("Call{0:03d}".format(i), sim=self.sim)self.sim.activate(j, j.execute(cell))yield hold, self, ran.expovariate(lam)

class Call(Process):""" Calls arrive at random at the cellphone hub"""

def execute(self, cell):self.trace("arrived")if cell.Nfree == 0:

self.trace("blocked and left")else:

self.trace("got a channel")cell.Nfree -= 1if cell.Nfree == 0:

self.trace("start busy period======")cell.busyStartTime = self.sim.now()cell.totalBusyVisits += 1

yield hold, self, ran.expovariate(mu)self.trace("finished")if cell.Nfree == 0:

self.trace("end busy period++++++")cell.busyEndTime = self.sim.now()busy = self.sim.now() - cell.busyStartTimeself.trace(" busy = {0:9.4f}".format(busy))cell.totalBusyTime += busy

cell.Nfree += 1

def trace(self, message):if TRACING:

print("{0:7.4f} {1:13s} {2}".format(self.sim.now(), message, self.name))

class Cell:""" Holds global measurements"""Nfree = 0totalBusyTime = 0.0totalBusyVisits = 0result = ()

class Statistician(Process):""" observes the system at intervals """

def execute(self, Nperiods, obsPeriod, obsGap, cell):cell.busyEndTime = self.sim.now() # simulation start timeif STRACING:

print("Busy time Number")for i in range(Nperiods):

yield hold, self, obsGap



cell.totalBusyTime = 0.0cell.totalBusyVisits = 0if cell.Nfree == 0:

cell.busyStartTime = self.sim.now()yield hold, self, obsPeriodif cell.Nfree == 0:

cell.totalBusyTime += self.sim.now() - cell.busyStartTimeif STRACING:

print("{0:7.3f} {1:5d}".format(cell.totalBusyTime, cell.totalBusyVisits))

self.sim.m.tally(cell.totalBusyTime)self.sim.bn.tally(cell.totalBusyVisits)

self.sim.stopSimulation()cell.result = (self.sim.m.mean(), self.sim.m.var(),

self.sim.bn.mean(), self.sim.bn.var())

# Experiment data -------------------------

NChannels = 4 # number of channels in the cellmaxN = 10000ranSeed = 3333333lam = 1.0 # per minutemu = 0.6667 # per minuteNperiods = 10obsPeriod = 60.0 # minutesobsGap = 15.0 # gap between observation periods

TRACING = FalseSTRACING = True

# Model -----------------------------------

class CellphoneModel(Simulation):def run(self):

self.initialize()self.m = Monitor(sim=self)self.bn = Monitor(sim=self)ran.seed(ranSeed)self.cell = Cell() # the cellphone towerself.cell.Nfree = NChannelss = Statistician('Statistician', sim=self)self.activate(s, s.execute(Nperiods, obsPeriod, obsGap, self.cell))g = CallSource('CallSource', sim=self)self.activate(g, g.execute(maxN, lam, self.cell))self.simulate(until=10000.0)

# Experiment ------------------------------modl = CellphoneModel()modl.run()

# Output ----------------------------------print('cellphone')# input data:print("lambda mu s Nperiods obsPeriod obsGap")FMT = "{0:7.4f} {1:6.4f} {2:4d} {3:4d} {4:6.2f} {5:6.2f}"



print(FMT.format(lam, mu, NChannels, Nperiods, obsPeriod, obsGap))

sr = modl.cell.resultprint("Busy Time: mean = {0:6.3f} var= {1:6.3f}".format(sr[0], sr[1]))print("Busy Number: mean = {0:6.3f} var= {1:6.3f}".format(sr[2], sr[3]))

Busy time Number5.857 152.890 61.219 41.512 74.237 51.140 13.596 83.964 84.146 92.091 5

cellphonelambda mu s Nperiods obsPeriod obsGap1.0000 0.6667 4 10 60.00 15.00

Busy Time: mean = 3.065 var= 2.193Busy Number: mean = 6.800 var= 12.360

Computer CPU: centralserver.py, centralserver_OO.py

A primitive central-server model with a single CPU and a single disk. A fixed number of users send “tasks” to thesystem which are processed and sent back to the user who then thinks for a time before sending a task back. Servicetimes are exponential. This system can be solved analytically. (TV)

The user of each terminal thinks for a time (exponential, mean 100.0 sec) and then submits a task to the CPU with aservice time (exponential, mean 1.0 sec). The user then remains idle until the task completes service and returns tohim or her. The arriving tasks form a single FCFS queue in front of the CPU.

Upon leaving the CPU a task is either finished (probability 0.20) and returns to its user to begin another think time,or requires data from a disk drive (probability 0.8). If a task requires access to the disk, it joins a FCFS queue beforeservice (service time at the disk, exponential, mean 1.39 sec). When finished with the disk, a task returns to the CPUqueue again for another compute time (exp, mean 1.$ sec).

The objective is to measure the throughput of the CPU (tasks per second)

""" centralserver.py

A time-shared computer consists of a singlecentral processing unit (CPU) and a number ofterminals. The operator of each terminal `thinks'for a time (exponential, mean 100.0 sec) and thensubmits a task to the computer with a service time(exponential, mean 1.0 sec). The operator thenremains idle until the task completes service andreturns to him or her. The arriving tasks form asingle FCFS queue in front of the CPU.

Upon leaving the CPU a task is either finished(probability 0.20) and returns to its operatorto begin another `think' time, or requires datafrom a disk drive (probability 0.8). If a task



requires access to the disk, it joins a FCFS queuebefore service (service time at the disk,exponential, mean 1.39 sec). When finished withthe disk, a task returns to the CPU queue againfor another compute time (exp, mean 1.$ sec).

the objective is to measure the throughput ofthe CPU (tasks per second)"""from SimPy.Simulation import *# from SimPy.SimulationTrace import *import random as ran


class Task(Process):""" A computer task requires at leastone use of the CPU and possibly accesses to adisk drive."""completed = 0rate = 0.0

def execute(self, maxCompletions):while Task.completed < maxCompletions:

self.debug(" starts thinking")thinktime = ran.expovariate(1.0 / MeanThinkTime)yield hold, self, thinktimeself.debug(" request cpu")yield request, self, cpuself.debug(" got cpu")CPUtime = ran.expovariate(1.0 / MeanCPUTime)yield hold, self, CPUtimeyield release, self, cpuself.debug(" finish cpu")while ran.random() < pDisk:

self.debug(" request disk")yield request, self, diskself.debug(" got disk")disktime = ran.expovariate(1.0 / MeanDiskTime)yield hold, self, disktimeself.debug(" finish disk")yield release, self, diskself.debug(" request cpu")yield request, self, cpuself.debug(" got cpu")CPUtime = ran.expovariate(1.0 / MeanCPUTime)yield hold, self, CPUtimeyield release, self, cpu

Task.completed += 1self.debug(" completed {0:d} tasks".format(Task.completed))Task.rate = Task.completed / float(now())

def debug(self, message):FMT = "{0:9.3f} {1} {2}"if DEBUG:

print(FMT.format(now(), self.name, message))



# Model ------------------------------def main():

initialize()for i in range(Nterminals):

t = Task(name="task{0}".format(i))activate(t, t.execute(MaxCompletions))

simulate(until=MaxrunTime)return (now(), Task.rate)

# Experiment data -------------------------

cpu = Resource(name='cpu')disk = Resource(name='disk')Nterminals = 3 # Number of terminals = TaskspDisk = 0.8 # prob. of going to diskMeanThinkTime = 10.0 # secondsMeanCPUTime = 1.0 # secondsMeanDiskTime = 1.39 # seconds

ran.seed(111113333)MaxrunTime = 20000.0MaxCompletions = 100DEBUG = False

# Experiment

result = main()

# Analysis/output -------------------------

print('centralserver')print('{0:7.4f}: CPU rate = {1:7.4f} tasks per second'.format(

result[0], result[1]))

centralserver913.7400: CPU rate = 0.1116 tasks per second

OO version

""" centralserver.py

A time-shared computer consists of a singlecentral processing unit (CPU) and a number ofterminals. The operator of each terminal `thinks'for a time (exponential, mean 100.0 sec) and thensubmits a task to the computer with a service time(exponential, mean 1.0 sec). The operator thenremains idle until the task completes service andreturns to him or her. The arriving tasks form asingle FCFS queue in front of the CPU.

Upon leaving the CPU a task is either finished(probability 0.20) and returns to its operatorto begin another `think' time, or requires datafrom a disk drive (probability 0.8). If a task



requires access to the disk, it joins a FCFS queuebefore service (service time at the disk,exponential, mean 1.39 sec). When finished withthe disk, a task returns to the CPU queue againfor another compute time (exp, mean 1.$ sec).

the objective is to measure the throughput ofthe CPU (tasks per second)"""from SimPy.Simulation import *# from SimPy.SimulationTrace import *import random as ran


class Task(Process):""" A computer task requires at leastone use of the CPU and possibly accesses to adisk drive."""completed = 0rate = 0.0

def execute(self, maxCompletions):while Task.completed < MaxCompletions:

self.debug(" starts thinking")thinktime = ran.expovariate(1.0 / MeanThinkTime)yield hold, self, thinktimeself.debug(" request cpu")yield request, self, self.sim.cpuself.debug(" got cpu")CPUtime = ran.expovariate(1.0 / MeanCPUTime)yield hold, self, CPUtimeyield release, self, self.sim.cpuself.debug(" finish cpu")while ran.random() < pDisk:

self.debug(" request disk")yield request, self, self.sim.diskself.debug(" got disk")disktime = ran.expovariate(1.0 / MeanDiskTime)yield hold, self, disktimeself.debug(" finish disk")yield release, self, self.sim.diskself.debug(" request cpu")yield request, self, self.sim.cpuself.debug(" got cpu")CPUtime = ran.expovariate(1.0 / MeanCPUTime)yield hold, self, CPUtimeyield release, self, self.sim.cpu

Task.completed += 1self.debug(" completed {0:d} tasks".format(Task.completed))Task.rate = Task.completed / float(self.sim.now())

def debug(self, message):FMT = "({0:9.3f} {1} {2}"if DEBUG:

print(FMT.format(self.sim.now(), self.name, message))



# Model ------------------------------class CentralServerModel(Simulation):

def run(self):self.initialize()self.cpu = Resource(name='cpu', sim=self)self.disk = Resource(name='disk', sim=self)for i in range(Nterminals):

t = Task(name="task{0}".format(i), sim=self)self.activate(t, t.execute(MaxCompletions))

self.simulate(until=MaxrunTime)return (self.now(), Task.rate)

# Experiment data -------------------------Nterminals = 3 # Number of terminals = TaskspDisk = 0.8 # prob. of going to diskMeanThinkTime = 10.0 # secondsMeanCPUTime = 1.0 # secondsMeanDiskTime = 1.39 # seconds

ran.seed(111113333)MaxrunTime = 20000.0MaxCompletions = 100DEBUG = False

# Experiment

result = CentralServerModel().run()

# Analysis/output -------------------------

print('centralserver')print('{0:7.4f}: CPU rate = {1:7.4f} tasks per second'.format(

result[0], result[1]))

centralserver913.7400: CPU rate = 0.1116 tasks per second

Messages on a Jackson Network: jacksonnetwork.py, jacksonnetwork_OO.py

A Jackson network with 3 nodes, exponential service times and probability switching. The simulation measures thedelay for jobs moving through the system. (TV)

Messages arrive randomly at rate 1.5 per second at a communication network with 3 nodes (computers). Each com-puter (node) can queue messages.

"""jacksonnetwork.py

Messages arrive randomly at rate 1.5 per secondat a communication network with 3 nodes(computers). Each computer (node) can queuemessages. Service-times are exponential withmean m_i at node i. These values are given in



the column headed n_i in the table below. Oncompleting service at node i a message transfersto node j with probability p_ij and leaves thesystem with probability p_i3.

These transition probabilities are as follows:Node m_i p_i0 p_i1 p_i2 p_i3i (leave)0 1.0 0 0.5 0.5 01 2.0 0 0 0.8 0.22 1.0 0.2 0 0 0.8

Your task is to estimate(1) the average time taken for jobs going

through the system and(2) the average number of jobs in the system.

"""from SimPy.Simulation import *# from SimPy.SimulationTrace import *import random as ran


def choose2dA(i, P):""" return a random choice from a set j = 0..n-1

with probs held in list of lists P[j] (n by n)using row icall: next = choose2d(i,P)

"""U = ran.random()sumP = 0.0for j in range(len(P[i])): # j = 0..n-1

sumP += P[i][j]if U < sumP:

breakreturn(j)

class Msg(Process):"""a message."""noInSystem = 0

def execute(self, i):""" executing a message """startTime = now()Msg.noInSystem += 1# print("DEBUG noInSystm = ",Msg.noInSystem)NoInSystem.observe(Msg.noInSystem)self.trace("Arrived node {0}".format(i))while i != 3:

yield request, self, node[i]self.trace("Got node {0}".format(i))st = ran.expovariate(1.0 / mean[i])yield hold, self, styield release, self, node[i]self.trace("Finished with {0}".format(i))



i = choose2dA(i, P)self.trace("Transfer to {0}".format(i))

TimeInSystem.tally(now() - startTime)self.trace("leaving {0} {1} in system".format(i, Msg.noInSystem))Msg.noInSystem -= 1NoInSystem.accum(Msg.noInSystem)

def trace(self, message):if MTRACING:

print("{0:7.4f} {1:3s} {2:10s}".format(now(), self.name, message))

class MsgSource(Process):""" generates a sequence of msgs """

def execute(self, rate, maxN):self.count = 0 # hold number of messages generatedself.trace("starting MsgSource")while (self.count < maxN):

self.count += 1p = Msg("Message {0}".format(self.count))activate(p, p.execute(startNode))yield hold, self, ran.expovariate(rate)

self.trace("generator finished with {0} ========".format(self.count))

def trace(self, message):if GTRACING:


# Experiment data -------------------------

rate = 1.5 # arrivals per secondmaxNumber = 1000 # of MessagesGTRACING = False # tracing Messages Source?

startNode = 0 # Messages always enter at node 0ran.seed(77777)MTRACING = False # tracing Message action?

TimeInSystem = Monitor("time")NoInSystem = Monitor("Number")

node = [Resource(1), Resource(1), Resource(1)]

mean = [1.0, 2.0, 1.0] # service times, secondsP = [[0, 0.5, 0.5, 0], # transition matrix P_ij

[0, 0, 0.8, 0.2],[0.2, 0, 0, 0.8]]


initialize()g = MsgSource(name="MsgSource")activate(g, g.execute(rate, maxNumber))simulate(until=5000.0)

# Analysis/output -------------------------



print('jacksonnetwork')print("Mean number in system = {0:10.4f}".format(NoInSystem.timeAverage()))print("Mean delay in system = {0:10.4f}".format(TimeInSystem.mean()))print("Total time run = {0:10.4f}".format(now()))print("Total jobs arrived = {0:10d}".format(g.count))print("Total jobs completed = {0:10d}".format(TimeInSystem.count()))print("Average arrival rate = {0:10.4f}".format(g.count / now()))

jacksonnetworkMean number in system = 250.8179Mean delay in system = 310.7710Total time run = 1239.0304Total jobs arrived = 1000Total jobs completed = 1000Average arrival rate = 0.8071

OO version

"""jacksonnetwork_OO.py

Messages arrive randomly at rate 1.5 per secondat a communication network with 3 nodes(computers). Each computer (node) can queuemessages. Service-times are exponential withmean m_i at node i. These values are given inthe column headed n_i in the table below. Oncompleting service at node i a message transfersto node j with probability p_ij and leaves thesystem with probability p_i3.

These transition probabilities are as follows:Node m_i p_i0 p_i1 p_i2 p_i3i (leave)0 1.0 0 0.5 0.5 01 2.0 0 0 0.8 0.22 1.0 0.2 0 0 0.8

Your task is to estimate(1) the average time taken for jobs going

through the system and(2) the average number of jobs in the system.



def choose2dA(i, P):""" return a random choice from a set j = 0..n-1

with probs held in list of lists P[j] (n by n)using row icall: next = choose2d(i,P)

"""U = ran.random()



sumP = 0.0for j in range(len(P[i])): # j = 0..n-1

sumP += P[i][j]if U < sumP:

breakreturn(j)

class Msg(Process):"""a message"""noInSystem = 0

def execute(self, i):""" executing a message """startTime = self.sim.now()Msg.noInSystem += 1# print("DEBUG noInSystm = ",Msg.noInSystem)self.sim.NoInSystem.observe(Msg.noInSystem)self.trace("Arrived node {0}".format(i))while i != 3:

yield request, self, self.sim.node[i]self.trace("Got node {0}".format(i))st = ran.expovariate(1.0 / mean[i])yield hold, self, styield release, self, self.sim.node[i]self.trace("Finished with {0}".format(i))i = choose2dA(i, P)self.trace("Transfer to {0}".format(i))

self.sim.TimeInSystem.tally(self.sim.now() - startTime)self.trace("leaving {0} {1} in system".format(i, Msg.noInSystem))Msg.noInSystem -= 1self.sim.NoInSystem.accum(Msg.noInSystem)

def trace(self, message):if MTRACING:

print("{0:7.4f} {1:3d} {2:10s}".format(self.sim.now(), self.name, message))

class MsgSource(Process):""" generates a sequence of msgs """

def execute(self, rate, maxN):self.count = 0 # hold number of messages generatedwhile (self.count < maxN):

self.count += 1p = Msg("Message {0}".format(self.count), sim=self.sim)self.sim.activate(p, p.execute(i=startNode))yield hold, self, ran.expovariate(rate)

self.trace("generator finished with {0} ========".format(self.count))

def trace(self, message):if GTRACING:


# Experiment data -------------------------



rate = 1.5 # arrivals per secondmaxNumber = 1000 # of MessagesGTRACING = False # tracing Messages Source?

startNode = 0 # Messages always enter at node 0ran.seed(77777)MTRACING = False # tracing Message action?

mean = [1.0, 2.0, 1.0] # service times, secondsP = [[0, 0.5, 0.5, 0], # transition matrix P_ij

[0, 0, 0.8, 0.2],[0.2, 0, 0, 0.8]]

# Model -----------------------------------

class JacksonnetworkModel(Simulation):def run(self):

self.initialize()self.TimeInSystem = Monitor("time", sim=self)self.NoInSystem = Monitor("Number", sim=self)self.node = [Resource(1, sim=self), Resource(

1, sim=self), Resource(1, sim=self)]self.g = MsgSource("MsgSource", sim=self)self.activate(self.g, self.g.execute(rate, maxNumber))self.simulate(until=5000.0)

# Experiment ------------------------------modl = JacksonnetworkModel()modl.run()

# Analysis/output -------------------------

print('jacksonnetwork')print("Mean number in system = {0:10.4f}".format(

modl.NoInSystem.timeAverage()))print("Mean delay in system = {0:10.4f}".format(modl.TimeInSystem.mean()))print("Total time run = {0:10.4f}".format(modl.now()))print("Total jobs arrived = {0:10d}".format(modl.g.count))print("Total jobs completed = {0:10d}".format(modl.TimeInSystem.count()))print("Average arrival rate = {0:10.4f}".format(modl.g.count / modl.now()))

jacksonnetworkMean number in system = 250.8179Mean delay in system = 310.7710Total time run = 1239.0304Total jobs arrived = 1000Total jobs completed = 1000Average arrival rate = 0.8071



Miscellaneous Models

Bank Customers who can renege: bank08renege.py, bank08renege_OO.py

(Note currently does not run under Python 3)

Use of reneging (compound yield request) based on bank08.py of the tutorial TheBank. Customers leave ifthey lose patience with waiting.

""" bank08

A counter with a random service timeand customers who renege. Based on the program bank08.pyfrom TheBank tutorial. (KGM)

"""from SimPy.Simulation import *from random import expovariate, seed, uniform


class Source(Process):""" Source generates customers randomly"""

def generate(self, number, interval, counter):for i in range(number):

c = Customer(name="Customer{0:02d}".format(i))activate(c, c.visit(counter, timeInBank=12.0))t = expovariate(1.0 / interval)yield hold, self, t

class Customer(Process):""" Customer arrives, is served and leaves """

def visit(self, counter, timeInBank=0):arrive = now()print("{0:7.4f} {1}: Here I am ".format(now(), self.name))

yield (request, self, counter), (hold, self, next(Customer.patience))

if self.acquired(counter):wait = now() - arriveprint("{0:7.4f} {1}: Waited {2:6.3f}".format(

now(), self.name, wait))tib = expovariate(1.0 / timeInBank)yield hold, self, tibyield release, self, counterprint("{0:7.4f} {1}: Finished".format(now(), self.name))

else:wait = now() - arriveprint("{0:7.4f} {1}: RENEGED after {2:6.3f}".format(

now(), self.name, wait))

def fpatience(minpatience=0, maxpatience=10000000000):while True:

yield uniform(minpatience, maxpatience)fpatience = staticmethod(fpatience)



# Model ---------------------------------------------

def model():counter = Resource(name="Karen")Customer.patience = Customer.fpatience(minpatience=1, maxpatience=3)initialize()source = Source('Source')activate(source, source.generate(NumCustomers,

interval=IntervalCustomers,counter=counter))

simulate(until=maxTime)

# Experiment data -------------------------

maxTime = 400.0theseed = 1234NumCustomers = 5IntervalCustomers = 10.0# Experiment ------------------------------

seed(theseed)print('bank08renege')model()

bank08renege0.0000 Customer00: Here I am0.0000 Customer00: Waited 0.0000.0902 Customer00: Finished

33.9482 Customer01: Here I am33.9482 Customer01: Waited 0.00044.4221 Customer01: Finished58.1367 Customer02: Here I am58.1367 Customer02: Waited 0.00069.2708 Customer03: Here I am70.3325 Customer03: RENEGED after 1.06271.9733 Customer04: Here I am73.6655 Customer04: RENEGED after 1.69275.5906 Customer02: Finished

OO version

""" bank08_OO

A counter with a random service timeand customers who renege. Based on the program bank08.pyfrom TheBank tutorial. (KGM)

"""from SimPy.Simulation import *from random import expovariate, seed, uniform





def generate(self, number, interval, counter):for i in range(number):

c = Customer(name="Customer{0:02d}".format(i), sim=self.sim)self.sim.activate(c, c.visit(counter, timeInBank=12.0))t = expovariate(1.0 / interval)yield hold, self, t


def visit(self, counter, timeInBank=0):arrive = self.sim.now()print("{0:7.4f} {1}: Here I am ".format(self.sim.now(), self.name))

yield (request, self, counter), (hold, self, next(Customer.patience))

if self.acquired(counter):wait = self.sim.now() - arriveprint("{0:7.4f} {1}: Waited {2:6.3f}".format(

self.sim.now(), self.name, wait))tib = expovariate(1.0 / timeInBank)yield hold, self, tibyield release, self, counterprint("{0:7.4f} {1}: Finished".format(self.sim.now(), self.name))

else:wait = self.sim.now() - arriveprint("{0:7.4f} {1}: RENEGED after {2:6.3f}".format(

self.sim.now(), self.name, wait))

def fpatience(minpatience=0, maxpatience=10000000000):while True:

yield uniform(minpatience, maxpatience)fpatience = staticmethod(fpatience)

# Model ---------------------------------------------

class BankModel(Simulation):def run(self):

self.initialize()counter = Resource(name="Karen", sim=self)Customer.patience = Customer.fpatience(minpatience=1, maxpatience=3)source = Source(name='Source', sim=self)self.activate(source, source.generate(NumCustomers,

interval=IntervalCustomers,counter=counter))

self.simulate(until=maxTime)

# Experiment data -------------------------

maxTime = 400.0theseed = 1234NumCustomers = 5IntervalCustomers = 10.0# Experiment ------------------------------



seed(theseed)print('bank08renege')BankModel().run()

bank08renege0.0000 Customer00: Here I am0.0000 Customer00: Waited 0.0000.0902 Customer00: Finished

33.9482 Customer01: Here I am33.9482 Customer01: Waited 0.00044.4221 Customer01: Finished58.1367 Customer02: Here I am58.1367 Customer02: Waited 0.00069.2708 Customer03: Here I am70.3325 Customer03: RENEGED after 1.06271.9733 Customer04: Here I am73.6655 Customer04: RENEGED after 1.69275.5906 Customer02: Finished

Carwash: Carwash.py, Carwash_OO.py

Using a Store object for implementing master/slave cooperation between processes. Scenario is a carwash installationwith multiple machines. Two model implementations are shown, one with the carwash as master in the cooperation,and the other with the car as master.

from SimPy.Simulation import *import random"""carwash.pyScenario:A carwash installation has nrMachines washing machines which wash a car inwashTime minutes.Cars arrive with a negative exponential interarrival time with a mean of tInterminutes.

Model the carwash operation as cooperation between two processes, the car beingwashed and the machine doing the washing.

Build two implementations:Model 1: the machine is master, the car slave;Model 2: the car is master, the machine is slave.

"""# Data:nrMachines = 2tInter = 2 # minuteswashtime = 3.5 # minutesinitialSeed = 123456simTime = 100

###################################################### Model 1: Carwash is master, car is slave#####################################################

class Carwash(Process):



"""Carwash machine; master"""

def __init__(self, name):Process.__init__(self, name)self.carBeingWashed = None


yield get, self, waitingCars, 1self.carBeingWashed = self.got[0]yield hold, self, washtimeself.carBeingWashed.doneSignal.signal(self.name)

class Car(Process):"""Car; slave"""

def __init__(self, name):Process.__init__(self, name)self.doneSignal = SimEvent()

def lifecycle(self):yield put, self, waitingCars, [self]yield waitevent, self, self.doneSignalwhichWash = self.doneSignal.signalparamprint("{0}: {1} done by {2}".format(now(), self.name, whichWash))

class CarGenerator(Process):"""Car arrival generation"""

def generate(self):i = 0while True:

yield hold, self, r.expovariate(1.0 / tInter)c = Car("car{0}".format(i))activate(c, c.lifecycle())i += 1

print("Model 1: carwash is master")print("--------------------------")initialize()r = random.Random()r.seed(initialSeed)waiting = []for j in range(1, 5):

c = Car("car-{0}".format(j))activate(c, c.lifecycle())waiting.append(c)

waitingCars = Store(capacity=40, initialBuffered=waiting)cw = []for i in range(2):

c = Carwash("Carwash {0}".format(i))cw.append(c)activate(c, c.lifecycle())

cg = CarGenerator()activate(cg, cg.generate())



simulate(until=simTime)print("waiting cars: {0}".format([x.name for x in waitingCars.theBuffer]))print("cars being washed: {0}".format([y.carBeingWashed.name for y in cw]))

###################################################### Model 2: Car is master, carwash is slave#####################################################

class CarM(Process):"""Car is master"""


def lifecycle(self):yield get, self, washers, 1whichWash = self.got[0]carsBeingWashed.append(self)yield hold, self, washtimeprint("{0}: {1} done by {2}".format(now(), self.name, whichWash.name))whichWash.doneSignal.signal()carsBeingWashed.remove(self)

class CarwashS(Process):def __init__(self, name):

Process.__init__(self, name)self.doneSignal = SimEvent()


yield put, self, washers, [self]yield waitevent, self, self.doneSignal

class CarGenerator1(Process):def generate(self):

i = 0while True:

yield hold, self, r.expovariate(1.0 / tInter)c = CarM("car{0}".format(i))activate(c, c.lifecycle())i += 1

print("\nModel 2: car is master")print("----------------------")initialize()r = random.Random()r.seed(initialSeed)washers = Store(capacity=nrMachines)carsBeingWashed = []for j in range(1, 5):

c = CarM("car-{0}".format(j))activate(c, c.lifecycle())

for i in range(2):cw = CarwashS("Carwash {0}".format(i))



activate(cw, cw.lifecycle())cg = CarGenerator1()activate(cg, cg.generate())simulate(until=simTime)print("waiting cars: {0}".format([x.name for x in washers.getQ]))print("cars being washed: {0}".format([x.name for x in carsBeingWashed]))

Model 1: carwash is master--------------------------3.5: car-1 done by Carwash 03.5: car-2 done by Carwash 17.0: car-3 done by Carwash 07.0: car-4 done by Carwash 117.5: car0 done by Carwash 017.5: car1 done by Carwash 121.0: car2 done by Carwash 021.0: car3 done by Carwash 124.5: car4 done by Carwash 024.5: car5 done by Carwash 128.0: car6 done by Carwash 028.0: car7 done by Carwash 131.5: car8 done by Carwash 031.5: car9 done by Carwash 135.0: car10 done by Carwash 035.0: car11 done by Carwash 138.5: car12 done by Carwash 038.5: car13 done by Carwash 142.0: car14 done by Carwash 042.0: car15 done by Carwash 145.5: car16 done by Carwash 045.5: car17 done by Carwash 149.0: car18 done by Carwash 049.0: car19 done by Carwash 152.5: car20 done by Carwash 052.5: car21 done by Carwash 156.0: car22 done by Carwash 056.0: car23 done by Carwash 159.5: car24 done by Carwash 059.5: car25 done by Carwash 163.0: car26 done by Carwash 063.0: car27 done by Carwash 166.5: car28 done by Carwash 066.5: car29 done by Carwash 170.0: car30 done by Carwash 070.0: car31 done by Carwash 173.5: car32 done by Carwash 073.5: car33 done by Carwash 177.0: car34 done by Carwash 077.0: car35 done by Carwash 180.5: car36 done by Carwash 080.5: car37 done by Carwash 184.0: car38 done by Carwash 084.0: car39 done by Carwash 187.5: car40 done by Carwash 087.5: car41 done by Carwash 191.0: car42 done by Carwash 091.0: car43 done by Carwash 194.5: car44 done by Carwash 0



94.5: car45 done by Carwash 198.0: car46 done by Carwash 098.0: car47 done by Carwash 1waiting cars: ['car50', 'car51', 'car52', 'car53', 'car54', 'car55', 'car56', 'car57',→˓ 'car58', 'car59']cars being washed: ['car48', 'car49']

Model 2: car is master----------------------3.5: car-1 done by Carwash 03.5: car-2 done by Carwash 17.0: car-3 done by Carwash 07.0: car-4 done by Carwash 110.5: car0 done by Carwash 010.5: car1 done by Carwash 114.0: car2 done by Carwash 014.0: car3 done by Carwash 117.5: car4 done by Carwash 017.5: car5 done by Carwash 121.0: car6 done by Carwash 021.0: car7 done by Carwash 124.5: car8 done by Carwash 024.5: car9 done by Carwash 128.0: car10 done by Carwash 028.0: car11 done by Carwash 131.5: car12 done by Carwash 031.5: car13 done by Carwash 135.0: car14 done by Carwash 035.0: car15 done by Carwash 138.5: car16 done by Carwash 038.5: car17 done by Carwash 142.0: car18 done by Carwash 042.0: car19 done by Carwash 145.5: car20 done by Carwash 045.5: car21 done by Carwash 149.0: car22 done by Carwash 049.0: car23 done by Carwash 152.5: car24 done by Carwash 052.5: car25 done by Carwash 156.0: car26 done by Carwash 056.0: car27 done by Carwash 159.5: car28 done by Carwash 059.5: car29 done by Carwash 163.0: car30 done by Carwash 063.0: car31 done by Carwash 166.5: car32 done by Carwash 066.5: car33 done by Carwash 170.0: car34 done by Carwash 070.0: car35 done by Carwash 173.5: car36 done by Carwash 073.5: car37 done by Carwash 177.0: car38 done by Carwash 077.0: car39 done by Carwash 180.5: car40 done by Carwash 080.5: car41 done by Carwash 184.0: car42 done by Carwash 084.0: car43 done by Carwash 187.5: car44 done by Carwash 0



87.5: car45 done by Carwash 191.0: car46 done by Carwash 091.0: car47 done by Carwash 194.5: car48 done by Carwash 094.5: car49 done by Carwash 198.0: car50 done by Carwash 098.0: car51 done by Carwash 1waiting cars: ['car54', 'car55', 'car56', 'car57', 'car58', 'car59']cars being washed: ['car52', 'car53']

Here is the OO version:

from SimPy.Simulation import *import random"""Carwash_OO.pyScenario:A carwash installation has nrMachines washing machines which wash a car inwashTime minutes.Cars arrive with a negative exponential interarrival time with a mean of tInterminutes.

Model the carwash operation as cooperation between two processes, the car beingwashed and the machine doing the washing.

Build two implementations:Model 1: the machine is master, the car slave;Model 2: the car is master, the machine is slave.

"""# Experiment data -------------------------nrMachines = 2tInter = 2 # minuteswashtime = 3.5 # minutesinitialSeed = 123456simTime = 100 # minutes

# Model 1 components ----------------------

class Carwash(Process):"""Carwash machine; master"""

def __init__(self, name, sim):Process.__init__(self, name=name, sim=sim)self.carBeingWashed = None


yield get, self, self.sim.waitingCars, 1self.carBeingWashed = self.got[0]yield hold, self, washtimeself.carBeingWashed.doneSignal.signal(self.name)

class Car(Process):"""Car; slave"""

def __init__(self, name, sim):



Process.__init__(self, name=name, sim=sim)self.doneSignal = SimEvent(sim=sim)

def lifecycle(self):yield put, self, self.sim.waitingCars, [self]yield waitevent, self, self.doneSignalwhichWash = self.doneSignal.signalparamprint("{0}: {1} done by {2}".format(

self.sim.now(), self.name, whichWash))

class CarGenerator(Process):"""Car arrival generation"""

def generate(self):i = 0while True:

yield hold, self, self.sim.r.expovariate(1.0 / tInter)c = Car("car{0}".format(i), sim=self.sim)self.sim.activate(c, c.lifecycle())i += 1

# Model 1: Carwash is master, car is slave

class CarWashModel1(Simulation):def run(self):

print("Model 1: carwash is master")print("--------------------------")self.initialize()self.r = random.Random()self.r.seed(initialSeed)waiting = []for j in range(1, 5):

c = Car("car{0}".format(-j), sim=self)self.activate(c, c.lifecycle())waiting.append(c)

self.waitingCars = Store(capacity=40, initialBuffered=waiting, sim=self)

cw = []for i in range(nrMachines):

c = Carwash("Carwash {0}".format(i), sim=self)cw.append(c)self.activate(c, c.lifecycle())

cg = CarGenerator(sim=self)self.activate(cg, cg.generate())self.simulate(until=simTime)

print("waiting cars: {0}".format([x.name for x in self.waitingCars.theBuffer]))

print("cars being washed: {0}".format([y.carBeingWashed.name for y in cw]))

# Experiment 1 ----------------------------CarWashModel1().run()

############################################




class CarM(Process):"""Car is master"""

def lifecycle(self):yield get, self, self.sim.washers, 1whichWash = self.got[0]self.sim.carsBeingWashed.append(self)yield hold, self, washtimeprint("{0}: {1} done by {2}".format(

self.sim.now(), self.name, whichWash.name))whichWash.doneSignal.signal()self.sim.carsBeingWashed.remove(self)

class CarwashS(Process):def __init__(self, name, sim):

Process.__init__(self, name=name, sim=sim)self.doneSignal = SimEvent(sim=sim)


yield put, self, self.sim.washers, [self]yield waitevent, self, self.doneSignal

class CarGenerator1(Process):def generate(self):

i = 0while True:

yield hold, self, self.sim.r.expovariate(1.0 / tInter)c = CarM("car{0}".format(i), sim=self.sim)self.sim.activate(c, c.lifecycle())i += 1

# Model 2: Car is master, carwash is slave

class CarWashModel2(Simulation):def run(self):

print("\nModel 2: car is master")print("----------------------")self.initialize()self.r = random.Random()self.r.seed(initialSeed)self.washers = Store(capacity=nrMachines, sim=self)self.carsBeingWashed = []for j in range(1, 5):

c = CarM("car{0}".format(-j), sim=self)self.activate(c, c.lifecycle())

for i in range(2):cw = CarwashS("Carwash {0}".format(i), sim=self)self.activate(cw, cw.lifecycle())

cg = CarGenerator1(sim=self)self.activate(cg, cg.generate())



self.simulate(until=simTime)

print("waiting cars: {0}".format([x.name for x in self.washers.getQ]))print("cars being washed: {0}".format(

[x.name for x in self.carsBeingWashed]))

# Experiment 1 ----------------------------CarWashModel2().run()

Model 1: carwash is master--------------------------3.5: car-1 done by Carwash 03.5: car-2 done by Carwash 17.0: car-3 done by Carwash 07.0: car-4 done by Carwash 117.5: car0 done by Carwash 017.5: car1 done by Carwash 121.0: car2 done by Carwash 021.0: car3 done by Carwash 124.5: car4 done by Carwash 024.5: car5 done by Carwash 128.0: car6 done by Carwash 028.0: car7 done by Carwash 131.5: car8 done by Carwash 031.5: car9 done by Carwash 135.0: car10 done by Carwash 035.0: car11 done by Carwash 138.5: car12 done by Carwash 038.5: car13 done by Carwash 142.0: car14 done by Carwash 042.0: car15 done by Carwash 145.5: car16 done by Carwash 045.5: car17 done by Carwash 149.0: car18 done by Carwash 049.0: car19 done by Carwash 152.5: car20 done by Carwash 052.5: car21 done by Carwash 156.0: car22 done by Carwash 056.0: car23 done by Carwash 159.5: car24 done by Carwash 059.5: car25 done by Carwash 163.0: car26 done by Carwash 063.0: car27 done by Carwash 166.5: car28 done by Carwash 066.5: car29 done by Carwash 170.0: car30 done by Carwash 070.0: car31 done by Carwash 173.5: car32 done by Carwash 073.5: car33 done by Carwash 177.0: car34 done by Carwash 077.0: car35 done by Carwash 180.5: car36 done by Carwash 080.5: car37 done by Carwash 184.0: car38 done by Carwash 084.0: car39 done by Carwash 187.5: car40 done by Carwash 087.5: car41 done by Carwash 1



91.0: car42 done by Carwash 091.0: car43 done by Carwash 194.5: car44 done by Carwash 094.5: car45 done by Carwash 198.0: car46 done by Carwash 098.0: car47 done by Carwash 1waiting cars: ['car50', 'car51', 'car52', 'car53', 'car54', 'car55', 'car56', 'car57',→˓ 'car58', 'car59']cars being washed: ['car48', 'car49']

Model 2: car is master----------------------3.5: car-1 done by Carwash 03.5: car-2 done by Carwash 17.0: car-3 done by Carwash 07.0: car-4 done by Carwash 110.5: car0 done by Carwash 010.5: car1 done by Carwash 114.0: car2 done by Carwash 014.0: car3 done by Carwash 117.5: car4 done by Carwash 017.5: car5 done by Carwash 121.0: car6 done by Carwash 021.0: car7 done by Carwash 124.5: car8 done by Carwash 024.5: car9 done by Carwash 128.0: car10 done by Carwash 028.0: car11 done by Carwash 131.5: car12 done by Carwash 031.5: car13 done by Carwash 135.0: car14 done by Carwash 035.0: car15 done by Carwash 138.5: car16 done by Carwash 038.5: car17 done by Carwash 142.0: car18 done by Carwash 042.0: car19 done by Carwash 145.5: car20 done by Carwash 045.5: car21 done by Carwash 149.0: car22 done by Carwash 049.0: car23 done by Carwash 152.5: car24 done by Carwash 052.5: car25 done by Carwash 156.0: car26 done by Carwash 056.0: car27 done by Carwash 159.5: car28 done by Carwash 059.5: car29 done by Carwash 163.0: car30 done by Carwash 063.0: car31 done by Carwash 166.5: car32 done by Carwash 066.5: car33 done by Carwash 170.0: car34 done by Carwash 070.0: car35 done by Carwash 173.5: car36 done by Carwash 073.5: car37 done by Carwash 177.0: car38 done by Carwash 077.0: car39 done by Carwash 180.5: car40 done by Carwash 080.5: car41 done by Carwash 1



84.0: car42 done by Carwash 084.0: car43 done by Carwash 187.5: car44 done by Carwash 087.5: car45 done by Carwash 191.0: car46 done by Carwash 091.0: car47 done by Carwash 194.5: car48 done by Carwash 094.5: car49 done by Carwash 198.0: car50 done by Carwash 098.0: car51 done by Carwash 1waiting cars: ['car54', 'car55', 'car56', 'car57', 'car58', 'car59']cars being washed: ['car52', 'car53']

Game of Life: CellularAutomata.py

A two-dimensional cellular automaton. Does the game of Life. (KGM)

from SimPy.Simulation import *"""CellularAutomata.pySimulation of two-dimensional cellular automata. Plays game of Life."""

class Autom(Process):def __init__(self, coords):

Process.__init__(self)self.x = coords[0]self.y = coords[1]self.state = False

def nrActiveNeighbours(self, x, y):nr = 0coords = [(xco + x, yco + y) for xco in (-1, 0, 1)

for yco in (-1, 0, 1) if not (xco == 0 and yco == 0)]

for a_coord in coords:try:

if cells[a_coord].state:nr += 1

except KeyError:# wrap aroundnux = divmod(a_coord[0], size)[1]nuy = divmod(a_coord[1], size)[1]if cells[(nux, nuy)].state:

nr += 1return nr

def decide(self, nrActive):return (self.state and (nrActive == 2 or nrActive == 3) or

(nrActive == 3))

def celllife(self):while True:

# calculate next statetemp = self.decide(self.nrActiveNeighbours(self.x, self.y))yield hold, self, 0.5



# set next stateself.state = tempyield hold, self, 0.5

class Show(Process):def __init__(self):


def picture(self):while True:

print("Generation {0}".format(now()))for i in range(size):

cls = " "for j in range(size):

if cells[(i, j)].state:cls += " *"

else:cls += " ."

print(cls)print("")yield hold, self, 1

size = 20cells = {}initialize()for i in range(size):

for j in range(size):a = cells[(i, j)] = Autom((i, j))activate(a, a.celllife())

# R-pentominocells[(9, 3)].state = Truecells[(10, 3)].state = Truecells[(9, 4)].state = Truecells[(8, 4)].state = Truecells[(9, 5)].state = True

cells[(5, 5)].state = Truecells[(5, 6)].state = Truecells[(4, 5)].state = Truecells[(4, 6)].state = Truecells[(4, 7)].state = Truecells[(10, 10)].state = Truecells[(10, 11)].state = Truecells[(10, 12)].state = Truecells[(10, 13)].state = Truecells[(11, 10)].state = Truecells[(11, 11)].state = Truecells[(11, 12)].state = Truecells[(11, 13)].state = True

print('CellularAutomata')s = Show()whenToStartShowing = 10activate(s, s.picture(), delay=whenToStartShowing)nrGenerations = 30



simulate(until=nrGenerations)

CellularAutomataGeneration 10

. . . . . . . . . . . . . . . . . . . .

. . . . . . . . . . . . . . . . . . . .

. . . . . . . . . . . . . . . . . . . .

. . . . . . . . . . . . . . . . . . . .

. . . . . . . . . . . . . . . . . . . .

. . . * . . . . . . . . . . . . . . . .

. . * * * . . . . . . . . . . . . . . .

. . * . * * . . . . . . . . . . . . . .

. . . . . * . . . . . . . . . . . . . .

. . . . . * * * . . . * * . . . . . . .

. . . . . . . . . . * . . * . . . . . .

. . . . . . . . . . * . . * . . . . . .

. . . . . . . . . . . * * . . . . . . .

. . . . . . . . . . . . . . . . . . . .

. . . . . . . . . . . . . . . . . . . .

. . . . . . . . . . . . . . . . . . . .

. . . . . . . . . . . . . . . . . . . .

. . . . . . . . . . . . . . . . . . . .

. . . . . . . . . . . . . . . . . . . .

. . . . . . . . . . . . . . . . . . . .

Generation 11. . . . . . . . . . . . . . . . . . . .. . . . . . . . . . . . . . . . . . . .. . . . . . . . . . . . . . . . . . . .. . . . . . . . . . . . . . . . . . . .. . . . . . . . . . . . . . . . . . . .. . * * * . . . . . . . . . . . . . . .. . * . . * . . . . . . . . . . . . . .. . * . . * . . . . . . . . . . . . . .. . . . . . . . . . . . . . . . . . . .. . . . . * * . . . . * * . . . . . . .. . . . . . * . . . * . . * . . . . . .. . . . . . . . . . * . . * . . . . . .. . . . . . . . . . . * * . . . . . . .. . . . . . . . . . . . . . . . . . . .. . . . . . . . . . . . . . . . . . . .. . . . . . . . . . . . . . . . . . . .. . . . . . . . . . . . . . . . . . . .. . . . . . . . . . . . . . . . . . . .. . . . . . . . . . . . . . . . . . . .. . . . . . . . . . . . . . . . . . . .

Generation 12. . . . . . . . . . . . . . . . . . . .. . . . . . . . . . . . . . . . . . . .. . . . . . . . . . . . . . . . . . . .. . . . . . . . . . . . . . . . . . . .. . . * . . . . . . . . . . . . . . . .. . * * * . . . . . . . . . . . . . . .. * * . . * . . . . . . . . . . . . . .. . . . . . . . . . . . . . . . . . . .. . . . . * * . . . . . . . . . . . . .. . . . . * * . . . . * * . . . . . . .



. . . . . * * . . . * . . * . . . . . .

. . . . . . . . . . * . . * . . . . . .

. . . . . . . . . . . * * . . . . . . .

. . . . . . . . . . . . . . . . . . . .

. . . . . . . . . . . . . . . . . . . .

. . . . . . . . . . . . . . . . . . . .

. . . . . . . . . . . . . . . . . . . .

. . . . . . . . . . . . . . . . . . . .

. . . . . . . . . . . . . . . . . . . .

. . . . . . . . . . . . . . . . . . . .

Generation 13. . . . . . . . . . . . . . . . . . . .. . . . . . . . . . . . . . . . . . . .. . . . . . . . . . . . . . . . . . . .. . . . . . . . . . . . . . . . . . . .. . * * * . . . . . . . . . . . . . . .. * . . * . . . . . . . . . . . . . . .. * * . * . . . . . . . . . . . . . . .. . . . . * * . . . . . . . . . . . . .. . . . . * * . . . . . . . . . . . . .. . . . * . . * . . . * * . . . . . . .. . . . . * * . . . * . . * . . . . . .. . . . . . . . . . * . . * . . . . . .. . . . . . . . . . . * * . . . . . . .. . . . . . . . . . . . . . . . . . . .. . . . . . . . . . . . . . . . . . . .. . . . . . . . . . . . . . . . . . . .. . . . . . . . . . . . . . . . . . . .. . . . . . . . . . . . . . . . . . . .. . . . . . . . . . . . . . . . . . . .. . . . . . . . . . . . . . . . . . . .

Generation 14. . . . . . . . . . . . . . . . . . . .. . . . . . . . . . . . . . . . . . . .. . . . . . . . . . . . . . . . . . . .. . . * . . . . . . . . . . . . . . . .. . * * * . . . . . . . . . . . . . . .. * . . * * . . . . . . . . . . . . . .. * * * * . . . . . . . . . . . . . . .. . . . * . * . . . . . . . . . . . . .. . . . * . . * . . . . . . . . . . . .. . . . * . . * . . . * * . . . . . . .. . . . . * * . . . * . . * . . . . . .. . . . . . . . . . * . . * . . . . . .. . . . . . . . . . . * * . . . . . . .. . . . . . . . . . . . . . . . . . . .. . . . . . . . . . . . . . . . . . . .. . . . . . . . . . . . . . . . . . . .. . . . . . . . . . . . . . . . . . . .. . . . . . . . . . . . . . . . . . . .. . . . . . . . . . . . . . . . . . . .. . . . . . . . . . . . . . . . . . . .

Generation 15. . . . . . . . . . . . . . . . . . . .. . . . . . . . . . . . . . . . . . . .



. . . . . . . . . . . . . . . . . . . .

. . * * * . . . . . . . . . . . . . . .

. . * . . * . . . . . . . . . . . . . .

. * . . . * . . . . . . . . . . . . . .

. * * . . . . . . . . . . . . . . . . .

. . * . * . . . . . . . . . . . . . . .

. . . * * . * * . . . . . . . . . . . .

. . . . * . . * . . . * * . . . . . . .

. . . . . * * . . . * . . * . . . . . .

. . . . . . . . . . * . . * . . . . . .

. . . . . . . . . . . * * . . . . . . .

. . . . . . . . . . . . . . . . . . . .

. . . . . . . . . . . . . . . . . . . .

. . . . . . . . . . . . . . . . . . . .

. . . . . . . . . . . . . . . . . . . .

. . . . . . . . . . . . . . . . . . . .

. . . . . . . . . . . . . . . . . . . .

. . . . . . . . . . . . . . . . . . . .

Generation 16. . . . . . . . . . . . . . . . . . . .. . . . . . . . . . . . . . . . . . . .. . . * . . . . . . . . . . . . . . . .. . * * * . . . . . . . . . . . . . . .. * * . . * . . . . . . . . . . . . . .. * . . . . . . . . . . . . . . . . . .. * * * . . . . . . . . . . . . . . . .. * * . * * . . . . . . . . . . . . . .. . . . * . * * . . . . . . . . . . . .. . . * * . . * . . . * * . . . . . . .. . . . . * * . . . * . . * . . . . . .. . . . . . . . . . * . . * . . . . . .. . . . . . . . . . . * * . . . . . . .. . . . . . . . . . . . . . . . . . . .. . . . . . . . . . . . . . . . . . . .. . . . . . . . . . . . . . . . . . . .. . . . . . . . . . . . . . . . . . . .. . . . . . . . . . . . . . . . . . . .. . . . . . . . . . . . . . . . . . . .. . . . . . . . . . . . . . . . . . . .

Generation 17. . . . . . . . . . . . . . . . . . . .. . . . . . . . . . . . . . . . . . . .. . * * * . . . . . . . . . . . . . . .. * . . * . . . . . . . . . . . . . . .. * . . * . . . . . . . . . . . . . . .

* . . * . . . . . . . . . . . . . . . .

* . . * * . . . . . . . . . . . . . . .. * . . * * * . . . . . . . . . . . . .. . * . . . * * . . . . . . . . . . . .. . . * * . . * . . . * * . . . . . . .. . . . * * * . . . * . . * . . . . . .. . . . . . . . . . * . . * . . . . . .. . . . . . . . . . . * * . . . . . . .. . . . . . . . . . . . . . . . . . . .. . . . . . . . . . . . . . . . . . . .. . . . . . . . . . . . . . . . . . . .



. . . . . . . . . . . . . . . . . . . .

. . . . . . . . . . . . . . . . . . . .

. . . . . . . . . . . . . . . . . . . .

. . . . . . . . . . . . . . . . . . . .

Generation 18. . . . . . . . . . . . . . . . . . . .. . . * . . . . . . . . . . . . . . . .. . * * * . . . . . . . . . . . . . . .. * . . * * . . . . . . . . . . . . . .

* * * * * . . . . . . . . . . . . . . .

* * * * . . . . . . . . . . . . . . . .

* * * * . . . . . . . . . . . . . . . .. * * . * . * * . . . . . . . . . . . .. . * . . . . * . . . . . . . . . . . .. . . * * . . * . . . * * . . . . . . .. . . * * * * . . . * . . * . . . . . .. . . . . * . . . . * . . * . . . . . .. . . . . . . . . . . * * . . . . . . .. . . . . . . . . . . . . . . . . . . .. . . . . . . . . . . . . . . . . . . .. . . . . . . . . . . . . . . . . . . .. . . . . . . . . . . . . . . . . . . .. . . . . . . . . . . . . . . . . . . .. . . . . . . . . . . . . . . . . . . .. . . . . . . . . . . . . . . . . . . .

Generation 19. . . . . . . . . . . . . . . . . . . .. . * * * . . . . . . . . . . . . . . .. . * . . * . . . . . . . . . . . . . .

* . . . . * . . . . . . . . . . . . . .. . . . . * . . . . . . . . . . . . . .. . . . . . . . . . . . . . . . . . . *. . . . * . . . . . . . . . . . . . . .

* . . . . . * * . . . . . . . . . . . .. * * . * * . * * . . . . . . . . . . .. . * . . . . * . . . * * . . . . . . .. . . * . . * . . . * . . * . . . . . .. . . . . * * . . . * . . * . . . . . .. . . . . . . . . . . * * . . . . . . .. . . . . . . . . . . . . . . . . . . .. . . . . . . . . . . . . . . . . . . .. . . . . . . . . . . . . . . . . . . .. . . . . . . . . . . . . . . . . . . .. . . . . . . . . . . . . . . . . . . .. . . . . . . . . . . . . . . . . . . .. . . . . . . . . . . . . . . . . . . .

Generation 20. . . * . . . . . . . . . . . . . . . .. . * * * . . . . . . . . . . . . . . .. * * . . * . . . . . . . . . . . . . .. . . . * * * . . . . . . . . . . . . .. . . . . . . . . . . . . . . . . . . .. . . . . . . . . . . . . . . . . . . .. . . . . . . . . . . . . . . . . . . .. * . * * . * * * . . . . . . . . . . .



. * * * . * . . * . . . . . . . . . . .

. * * . * * . * * . . * * . . . . . . .

. . . . . * * * . . * . . * . . . . . .

. . . . . * * . . . * . . * . . . . . .

. . . . . . . . . . . * * . . . . . . .

. . . . . . . . . . . . . . . . . . . .

. . . . . . . . . . . . . . . . . . . .

. . . . . . . . . . . . . . . . . . . .

. . . . . . . . . . . . . . . . . . . .

. . . . . . . . . . . . . . . . . . . .

. . . . . . . . . . . . . . . . . . . .

. . . . . . . . . . . . . . . . . . . .

Generation 21. . * * * . . . . . . . . . . . . . . .. * . . * . . . . . . . . . . . . . . .. * * . . . * . . . . . . . . . . . . .. . . . * * * . . . . . . . . . . . . .. . . . . * . . . . . . . . . . . . . .. . . . . . . . . . . . . . . . . . . .. . . . . . . * . . . . . . . . . . . .. * . * * * * * * . . . . . . . . . . .

* . . . . . . . . * . . . . . . . . . .. * . . . . . . * * . * * . . . . . . .. . . . . . . . * * * . . * . . . . . .. . . . . * . * . . * . . * . . . . . .. . . . . . . . . . . * * . . . . . . .. . . . . . . . . . . . . . . . . . . .. . . . . . . . . . . . . . . . . . . .. . . . . . . . . . . . . . . . . . . .. . . . . . . . . . . . . . . . . . . .. . . . . . . . . . . . . . . . . . . .. . . . . . . . . . . . . . . . . . . .. . . . . . . . . . . . . . . . . . . .

Generation 22. . * * * . . . . . . . . . . . . . . .. * . . * * . . . . . . . . . . . . . .. * * * * . * . . . . . . . . . . . . .. . . . * . * . . . . . . . . . . . . .. . . . * * * . . . . . . . . . . . . .. . . . . . . . . . . . . . . . . . . .. . . . * * . * * . . . . . . . . . . .. . . . * * * * * . . . . . . . . . . .

* * * . * * * . . * * . . . . . . . . .. . . . . . . . . . . * * . . . . . . .. . . . . . . * . . . . . * . . . . . .. . . . . . . . * . * . . * . . . . . .. . . . . . . . . . . * * . . . . . . .. . . . . . . . . . . . . . . . . . . .. . . . . . . . . . . . . . . . . . . .. . . . . . . . . . . . . . . . . . . .. . . . . . . . . . . . . . . . . . . .. . . . . . . . . . . . . . . . . . . .. . . . . . . . . . . . . . . . . . . .. . . * . . . . . . . . . . . . . . . .

Generation 23



. . * . . * . . . . . . . . . . . . . .

. * . . . . . . . . . . . . . . . . . .

. * * . . . * . . . . . . . . . . . . .

. . * . . . * * . . . . . . . . . . . .

. . . . * . * . . . . . . . . . . . . .

. . . . . . . * . . . . . . . . . . . .

. . . . * . . . * . . . . . . . . . . .

. * . . . . . . . . . . . . . . . . . .

. * . * * . . . * * * * . . . . . . . .

. * . . . * * . . . * * * . . . . . . .

. . . . . . . . . . . * . * . . . . . .

. . . . . . . . . . . * . * . . . . . .

. . . . . . . . . . . * * . . . . . . .

. . . . . . . . . . . . . . . . . . . .

. . . . . . . . . . . . . . . . . . . .

. . . . . . . . . . . . . . . . . . . .

. . . . . . . . . . . . . . . . . . . .

. . . . . . . . . . . . . . . . . . . .

. . . . . . . . . . . . . . . . . . . .

. . * * * . . . . . . . . . . . . . . .

Generation 24. * * . * . . . . . . . . . . . . . . .. * . . . . . . . . . . . . . . . . . .. * * . . . * * . . . . . . . . . . . .. * * * . . * * . . . . . . . . . . . .. . . . . * * . . . . . . . . . . . . .. . . . . * . * . . . . . . . . . . . .. . . . . . . . . . . . . . . . . . . .. . * * * . . . * . * . . . . . . . . .

* * . . * * . . . * . . * . . . . . . .. . * . * * . . . . . . . . . . . . . .. . . . . . . . . . . . . * . . . . . .. . . . . . . . . . * * . * . . . . . .. . . . . . . . . . . * * . . . . . . .. . . . . . . . . . . . . . . . . . . .. . . . . . . . . . . . . . . . . . . .. . . . . . . . . . . . . . . . . . . .. . . . . . . . . . . . . . . . . . . .. . . . . . . . . . . . . . . . . . . .. . . * . . . . . . . . . . . . . . . .. . * * * . . . . . . . . . . . . . . .

Generation 25. * . . * . . . . . . . . . . . . . . .

* . . * . . . . . . . . . . . . . . . .

* . . * . . * * . . . . . . . . . . . .. * . * . . . . . . . . . . . . . . . .. . * . * * . . . . . . . . . . . . . .. . . . . * . . . . . . . . . . . . . .. . . * * . . . . . . . . . . . . . . .. * * * * * . . . * . . . . . . . . . .. * . . . . . . . * . . . . . . . . . .. * . * * * . . . . . . . . . . . . . .. . . . . . . . . . . . * . . . . . . .. . . . . . . . . . * * . * . . . . . .. . . . . . . . . . * * * . . . . . . .. . . . . . . . . . . . . . . . . . . .



. . . . . . . . . . . . . . . . . . . .

. . . . . . . . . . . . . . . . . . . .

. . . . . . . . . . . . . . . . . . . .

. . . . . . . . . . . . . . . . . . . .

. . * * * . . . . . . . . . . . . . . .

. * . . * . . . . . . . . . . . . . . .

Generation 26

* * * * * . . . . . . . . . . . . . . .

* * * * * . . . . . . . . . . . . . . .

* * . * * . . . . . . . . . . . . . . .. * . * . * * . . . . . . . . . . . . .. . * * * * . . . . . . . . . . . . . .. . . . . * . . . . . . . . . . . . . .. . . . . . . . . . . . . . . . . . . .. * . . . * . . . . . . . . . . . . . .

* * . . . . . . . . . . . . . . . . . .. . * . * . . . . . . . . . . . . . . .. . . . * . . . . . . * * . . . . . . .. . . . . . . . . . * . . * . . . . . .. . . . . . . . . . * . * . . . . . . .. . . . . . . . . . . * . . . . . . . .. . . . . . . . . . . . . . . . . . . .. . . . . . . . . . . . . . . . . . . .. . . . . . . . . . . . . . . . . . . .. . . * . . . . . . . . . . . . . . . .. . * * * . . . . . . . . . . . . . . .. * . . * * . . . . . . . . . . . . . .

Generation 27. . . . . . . . . . . . . . . . . . . .. . . . . * . . . . . . . . . . . . . *. . . . . . . . . . . . . . . . . . . .

* * . . . . * . . . . . . . . . . . . .. . * * . . . . . . . . . . . . . . . .. . . * . * . . . . . . . . . . . . . .. . . . . . . . . . . . . . . . . . . .

* * . . . . . . . . . . . . . . . . . .

* * * . . . . . . . . . . . . . . . . .. * . * . . . . . . . . . . . . . . . .. . . * . . . . . . . * * . . . . . . .. . . . . . . . . . * . . * . . . . . .. . . . . . . . . . * . * . . . . . . .. . . . . . . . . . . * . . . . . . . .. . . . . . . . . . . . . . . . . . . .. . . . . . . . . . . . . . . . . . . .. . . . . . . . . . . . . . . . . . . .. . * * * . . . . . . . . . . . . . . .. . * . . * . . . . . . . . . . . . . .

* . . . . * . . . . . . . . . . . . . .

Generation 28. . . . . . . . . . . . . . . . . . . .. . . . . . . . . . . . . . . . . . . .

* . . . . . . . . . . . . . . . . . . .. * * . . . . . . . . . . . . . . . . .. * * * * . . . . . . . . . . . . . . .. . * * * . . . . . . . . . . . . . . .



. . . . . . . . . . . . . . . . . . . .

* . * . . . . . . . . . . . . . . . . .. . . . . . . . . . . . . . . . . . . .

* * . * . . . . . . . . . . . . . . . .. . * . . . . . . . . * * . . . . . . .. . . . . . . . . . * . . * . . . . . .. . . . . . . . . . * . * . . . . . . .. . . . . . . . . . . * . . . . . . . .. . . . . . . . . . . . . . . . . . . .. . . . . . . . . . . . . . . . . . . .. . . * . . . . . . . . . . . . . . . .. . * * * . . . . . . . . . . . . . . .. * * . . * . . . . . . . . . . . . . .. . . . . . . . . . . . . . . . . . . .

Generation 29. . . . . . . . . . . . . . . . . . . .. . . . . . . . . . . . . . . . . . . .. * . . . . . . . . . . . . . . . . . .

* . . . . . . . . . . . . . . . . . . .. . . . * . . . . . . . . . . . . . . .. * . . * . . . . . . . . . . . . . . .. * * . . . . . . . . . . . . . . . . .. . . . . . . . . . . . . . . . . . . .

* . * . . . . . . . . . . . . . . . . .. * * . . . . . . . . . . . . . . . . .. * * . . . . . . . . * * . . . . . . .. . . . . . . . . . * . . * . . . . . .. . . . . . . . . . * . * . . . . . . .. . . . . . . . . . . * . . . . . . . .. . . . . . . . . . . . . . . . . . . .. . . . . . . . . . . . . . . . . . . .. . * * * . . . . . . . . . . . . . . .. * . . * . . . . . . . . . . . . . . .. * * . * . . . . . . . . . . . . . . .. . . . . . . . . . . . . . . . . . . .

Generation 30. . . . . . . . . . . . . . . . . . . .. . . . . . . . . . . . . . . . . . . .. . . . . . . . . . . . . . . . . . . .. . . . . . . . . . . . . . . . . . . .. . . . . . . . . . . . . . . . . . . .. * * * . . . . . . . . . . . . . . . .. * * . . . . . . . . . . . . . . . . .. . * . . . . . . . . . . . . . . . . .. . * . . . . . . . . . . . . . . . . .

* . . * . . . . . . . . . . . . . . . .. * * . . . . . . . . * * . . . . . . .. . . . . . . . . . * . . * . . . . . .. . . . . . . . . . * . * . . . . . . .. . . . . . . . . . . * . . . . . . . .. . . . . . . . . . . . . . . . . . . .. . . * . . . . . . . . . . . . . . . .. . * * * . . . . . . . . . . . . . . .. * . . * * . . . . . . . . . . . . . .. * * * . . . . . . . . . . . . . . . .. . . . . . . . . . . . . . . . . . . .



SimPy’s event signalling synchronisation constructs: demoSimPyEvents.py

Demo of the event signalling constructs. Three small simulations are included: Pavlov’s drooling dogs, an activitysimulation where a job is completed after a number of parallel activities, and the simulation of a US-style 4-way stopintersection.

from SimPy.Simulation import *import random"""

Demo of SimPy's event signalling synchronization constructs"""

print('demoSimPyEvents')

# Pavlov's dogs"""Scenario:Dogs start to drool when Pavlov rings the bell."""

class BellMan(Process):def ring(self):

while True:bell.signal()print("{0} {1} rings bell".format(now(), self.name))yield hold, self, 5

class PavlovDog(Process):def behave(self):

while True:yield waitevent, self, bellprint("{0} {1} drools".format(now(), self.name))

random.seed(111333555)initialize()bell = SimEvent("bell")for i in range(4):

p = PavlovDog("Dog {0}".format(i + 1))activate(p, p.behave())

b = BellMan("Pavlov")activate(b, b.ring())print("\n Pavlov's dogs")simulate(until=10)

# PERT simulation"""Scenario:A job (TotalJob) requires 10 parallel activities with random duration to becompleted."""



class Activity(Process):def __init__(self, name):

Process.__init__(self, name)self.event = SimEvent("completion of {0}".format(self.name))allEvents.append(self.event)

def perform(self):yield hold, self, random.randint(1, 100)self.event.signal()print("{0} Event '{1}' fired".format(now(), self.event.name))

class TotalJob(Process):def perform(self, allEvents):

for e in allEvents:yield waitevent, self, e

print(now(), "All done")

random.seed(111333555)initialize()allEvents = []for i in range(10):

a = Activity("Activity {0}".format(i + 1))activate(a, a.perform())

t = TotalJob()activate(t, t.perform(allEvents))print("\n PERT network simulation")simulate(until=100)

# US-style4-way stop intersection"""Scenario:At a US-style 4-way stop intersection, a car may only enter the intersectionwhen it is free.Cars enter in FIFO manner."""

class Car(Process):def drive(self):

print("{0:4.1f} {1} waiting to enter intersection".format(now(),self.name))

yield queueevent, self, intersectionFree# Intersection free, enter . .# Begin Critical Sectionyield hold, self, 1 # drive acrossprint("{0:4.1f} {1} crossed intersection".format(now(), self.name))# End Critical SectionintersectionFree.signal()

random.seed(111333555)initialize()intersectionFree = SimEvent("Intersection free")intersectionFree.signal()arrtime = 0.0for i in range(20):



c = Car("Car {0}".format(i + 1))activate(c, c.drive(), at=arrtime)arrtime += 0.2

print("\n 4-way stop intersection")print(simulate(until=100))

demoSimPyEvents

Pavlov's dogs0 Pavlov rings bell0 Dog 4 drools0 Dog 3 drools0 Dog 2 drools0 Dog 1 drools5 Pavlov rings bell5 Dog 1 drools5 Dog 2 drools5 Dog 3 drools5 Dog 4 drools10 Pavlov rings bell10 Dog 4 drools10 Dog 3 drools10 Dog 2 drools10 Dog 1 drools

PERT network simulation13 Event 'completion of Activity 6' fired19 Event 'completion of Activity 10' fired22 Event 'completion of Activity 7' fired42 Event 'completion of Activity 1' fired48 Event 'completion of Activity 4' fired55 Event 'completion of Activity 8' fired69 Event 'completion of Activity 5' fired83 Event 'completion of Activity 2' fired90 Event 'completion of Activity 9' fired93 Event 'completion of Activity 3' fired93 All done

4-way stop intersection0.0 Car 1 waiting to enter intersection0.2 Car 2 waiting to enter intersection0.4 Car 3 waiting to enter intersection0.6 Car 4 waiting to enter intersection0.8 Car 5 waiting to enter intersection1.0 Car 6 waiting to enter intersection1.0 Car 1 crossed intersection1.2 Car 7 waiting to enter intersection1.4 Car 8 waiting to enter intersection1.6 Car 9 waiting to enter intersection1.8 Car 10 waiting to enter intersection2.0 Car 11 waiting to enter intersection2.0 Car 2 crossed intersection2.2 Car 12 waiting to enter intersection2.4 Car 13 waiting to enter intersection2.6 Car 14 waiting to enter intersection2.8 Car 15 waiting to enter intersection3.0 Car 3 crossed intersection3.0 Car 16 waiting to enter intersection



3.2 Car 17 waiting to enter intersection3.4 Car 18 waiting to enter intersection3.6 Car 19 waiting to enter intersection3.8 Car 20 waiting to enter intersection4.0 Car 4 crossed intersection5.0 Car 5 crossed intersection6.0 Car 6 crossed intersection7.0 Car 7 crossed intersection8.0 Car 8 crossed intersection9.0 Car 9 crossed intersection

10.0 Car 10 crossed intersection11.0 Car 11 crossed intersection12.0 Car 12 crossed intersection13.0 Car 13 crossed intersection14.0 Car 14 crossed intersection15.0 Car 15 crossed intersection16.0 Car 16 crossed intersection17.0 Car 17 crossed intersection18.0 Car 18 crossed intersection19.0 Car 19 crossed intersection20.0 Car 20 crossed intersectionSimPy: No more events at time 20

Find the Shortest Path: shortestPath_SimPy.py, shortestPath_SimPy_OO.py

A fun example of using SimPy for non-queuing work. It simulates a searcher through a graph, seeking the shortestpath. (KGM)

from SimPy.Simulation import *""" shortestPath_SimPy.py

Finds the shortest path in a network.Author: Klaus Muller

"""

class node:def __init__(self):

self.reached = 0

class searcher(Process):def __init__(self, graph, path, length, from_node, to_node,

distance, goal_node):Process.__init__(self)self.path = path[:]self.length = lengthself.from_node = from_nodeself.to_node = to_nodeself.distance = distanceself.graph = graphself.goal_node = goal_node

def run(self, to_node):if DEMO:

print("Path so far: {0} (length {1}). "



"Search from {2} to {3}".format(self.path, self.length,self.from_node, to_node))

yield hold, self, self.distanceif not nodes[to_node].reached:

self.path.append(to_node)self.length += self.distancenodes[to_node].reached = 1if to_node == self.goal_node:

print("SHORTEST PATH", self.path, "Length:", self.length)stopSimulation()

else:for i in self.graph[to_node]:

s = searcher(graph=self.graph, path=self.path,length=self.length, from_node=i[0],to_node=i[1], distance=i[2],goal_node=self.goal_node)

activate(s, s.run(i[1]))

print('shortestPath_SimPy')initialize()nodes = {}DEMO = 1for i in ("Atown", "Btown", "Ccity", "Dpueblo", "Evillage", "Fstadt"):

nodes[i] = node()""" Format graph definition:a_graph={node_id:[(from,to,distance),

(from,to,distance)],node_id:[ . . . ])

"""net = {"Atown": (("Atown", "Btown", 3.5), ("Atown", "Ccity", 1),

("Atown", "Atown", 9), ("Atown", "Evillage", 0.5)),"Btown": (("Btown", "Ccity", 5),),"Ccity": (("Ccity", "Ccity", 1), ("Ccity", "Fstadt", 9),

("Ccity", "Dpueblo", 3), ("Ccity", "Atown", 3)),"Dpueblo": (("Dpueblo", "Ccity", 2), ("Dpueblo", "Fstadt", 10)),"Evillage": (("Evillage", "Btown", 1),),"Fstadt": (("Fstadt", "Ccity", 3),)}

if DEMO:print("Search for shortest path from {0} to {1} \nin graph {2}".format(

"Atown", "Fstadt", sorted(net.items())))startup = searcher(graph=net, path=[], length=0, from_node="Atown",

to_node="Atown", distance=0, goal_node="Fstadt")activate(startup, startup.run("Atown"))simulate(until=10000)

shortestPath_SimPySearch for shortest path from Atown to Fstadtin graph [('Atown', (('Atown', 'Btown', 3.5), ('Atown', 'Ccity', 1), ('Atown', 'Atown→˓', 9), ('Atown', 'Evillage', 0.5))), ('Btown', (('Btown', 'Ccity', 5),)), ('Ccity',→˓(('Ccity', 'Ccity', 1), ('Ccity', 'Fstadt', 9), ('Ccity', 'Dpueblo', 3), ('Ccity',→˓'Atown', 3))), ('Dpueblo', (('Dpueblo', 'Ccity', 2), ('Dpueblo', 'Fstadt', 10))), (→˓'Evillage', (('Evillage', 'Btown', 1),)), ('Fstadt', (('Fstadt', 'Ccity', 3),))]Path so far: [] (length 0). Search from Atown to AtownPath so far: ['Atown'] (length 0). Search from Atown to BtownPath so far: ['Atown'] (length 0). Search from Atown to CcityPath so far: ['Atown'] (length 0). Search from Atown to AtownPath so far: ['Atown'] (length 0). Search from Atown to Evillage



Path so far: ['Atown', 'Evillage'] (length 0.5). Search from Evillage to BtownPath so far: ['Atown', 'Ccity'] (length 1). Search from Ccity to CcityPath so far: ['Atown', 'Ccity'] (length 1). Search from Ccity to FstadtPath so far: ['Atown', 'Ccity'] (length 1). Search from Ccity to DpuebloPath so far: ['Atown', 'Ccity'] (length 1). Search from Ccity to AtownPath so far: ['Atown', 'Evillage', 'Btown'] (length 1.5). Search from Btown to CcityPath so far: ['Atown', 'Ccity', 'Dpueblo'] (length 4). Search from Dpueblo to CcityPath so far: ['Atown', 'Ccity', 'Dpueblo'] (length 4). Search from Dpueblo to FstadtSHORTEST PATH ['Atown', 'Ccity', 'Fstadt'] Length: 10


from SimPy.Simulation import *""" shortestPath_SimPy_OO.py

Finds the shortest path in a network.Author: Klaus Muller

"""# Model components ------------------------

class node:def __init__(self):

self.reached = 0

class searcher(Process):def __init__(self, graph, path, length, from_node,

to_node, distance, goal_node, sim):Process.__init__(self, sim=sim)self.path = path[:]self.length = lengthself.from_node = from_nodeself.to_node = to_nodeself.distance = distanceself.graph = graphself.goal_node = goal_node

def run(self, to_node):if DEMO:

print("Path so far: {0} (length {1}). ""Search from {2} to {3}".format(self.path,

self.length,self.from_node,to_node))

yield hold, self, self.distanceif not self.sim.nodes[to_node].reached:

self.path.append(to_node)self.length += self.distanceself.sim.nodes[to_node].reached = 1if to_node == self.goal_node:

print("SHORTEST PATH", self.path, "Length:", self.length)self.sim.stopSimulation()

else:for i in self.graph[to_node]:

s = searcher(graph=self.graph, path=self.path,length=self.length, from_node=i[0],to_node=i[1], distance=i[2],



goal_node=self.goal_node, sim=self.sim)self.sim.activate(s, s.run(i[1]))

# Model -----------------------------------

class ShortestPathModel(Simulation):def search(self):

print('shortestPath_SimPy')self.initialize()self.nodes = {}for i in ("Atown", "Btown", "Ccity", "Dpueblo", "Evillage", "Fstadt"):

self.nodes[i] = node()""" Format graph definition:a_graph={node_id:[(from,to,distance),

(from,to,distance)],node_id:[ . . . ])

"""net = {"Atown": (("Atown", "Btown", 3.5), ("Atown", "Ccity", 1),

("Atown", "Atown", 9), ("Atown", "Evillage", 0.5)),"Btown": (("Btown", "Ccity", 5),),"Ccity": (("Ccity", "Ccity", 1), ("Ccity", "Fstadt", 9),

("Ccity", "Dpueblo", 3), ("Ccity", "Atown", 3)),"Dpueblo": (("Dpueblo", "Ccity", 2), ("Dpueblo", "Fstadt", 10)),"Evillage": (("Evillage", "Btown", 1),),"Fstadt": (("Fstadt", "Ccity", 3),)}

if DEMO:print("Search for shortest path from {0} to {1} \n"

"in graph {2}".format("Atown","Fstadt",sorted(net.items())))

startup = searcher(graph=net, path=[], length=0, from_node="Atown",to_node="Atown", distance=0, goal_node="Fstadt",sim=self)

self.activate(startup, startup.run("Atown"))self.simulate(until=10000)

# Experiment ------------------------------DEMO = 1ShortestPathModel().search()

shortestPath_SimPySearch for shortest path from Atown to Fstadtin graph [('Atown', (('Atown', 'Btown', 3.5), ('Atown', 'Ccity', 1), ('Atown', 'Atown→˓', 9), ('Atown', 'Evillage', 0.5))), ('Btown', (('Btown', 'Ccity', 5),)), ('Ccity',→˓(('Ccity', 'Ccity', 1), ('Ccity', 'Fstadt', 9), ('Ccity', 'Dpueblo', 3), ('Ccity',→˓'Atown', 3))), ('Dpueblo', (('Dpueblo', 'Ccity', 2), ('Dpueblo', 'Fstadt', 10))), (→˓'Evillage', (('Evillage', 'Btown', 1),)), ('Fstadt', (('Fstadt', 'Ccity', 3),))]Path so far: [] (length 0). Search from Atown to AtownPath so far: ['Atown'] (length 0). Search from Atown to BtownPath so far: ['Atown'] (length 0). Search from Atown to CcityPath so far: ['Atown'] (length 0). Search from Atown to AtownPath so far: ['Atown'] (length 0). Search from Atown to EvillagePath so far: ['Atown', 'Evillage'] (length 0.5). Search from Evillage to BtownPath so far: ['Atown', 'Ccity'] (length 1). Search from Ccity to CcityPath so far: ['Atown', 'Ccity'] (length 1). Search from Ccity to FstadtPath so far: ['Atown', 'Ccity'] (length 1). Search from Ccity to Dpueblo



Path so far: ['Atown', 'Ccity'] (length 1). Search from Ccity to AtownPath so far: ['Atown', 'Evillage', 'Btown'] (length 1.5). Search from Btown to CcityPath so far: ['Atown', 'Ccity', 'Dpueblo'] (length 4). Search from Dpueblo to CcityPath so far: ['Atown', 'Ccity', 'Dpueblo'] (length 4). Search from Dpueblo to FstadtSHORTEST PATH ['Atown', 'Ccity', 'Fstadt'] Length: 10

Machine Shop Model: Machineshop.py, Machineshop_OO.py

A workshop has n identical machines. A stream of jobs (enough to keep the machines busy) arrives. Each machinebreaks down periodically. Repairs are carried out by one repairman. The repairman has other, less important tasksto perform, too. Once he starts one of those, he completes it before starting with the machine repair. The workshopworks continuously.

This is an example of the use of the interrupt() method. (KGM)

from SimPy.Simulation import *import random"""Machineshop Model

An example showing interrupts and priority queuing.

Scenario:

A workshop has n identical machines. A stream of jobs (enough to keepthe machines busy) arrives. Each machine breaks downperiodically. Repairs are carried out by one repairman. The repairmanhas other, less important tasks to perform, too. Once he starts one ofthose, he completes it before starting with the machine repair. Theworkshop works continously. """


class Machine(Process):def __init__(self, name):

Process.__init__(self, name)myBreaker = Breakdown(self)activate(myBreaker, myBreaker.breakmachine())self.partsMade = 0

def working(self):while True:

yield hold, self, timePerPart()if self.interrupted():

# broken downparttimeleft = self.interruptLeftyield request, self, repairman, 1yield hold, self, repairtime# repairedyield release, self, repairmanyield hold, self, parttimeleft# part completedself.partsMade += 1

else:# part made



self.partsMade += 1

class Breakdown(Process):def __init__(self, myMachine):

Process.__init__(self)self.myMachine = myMachine

def breakmachine(self):while True:

yield hold, self, timeToFailure()self.interrupt(self.myMachine)

class OtherJobs(Process):def __init__(self):


def doingJobs(self):while True:

yield request, self, repairman, 0# starts working on jobsyield hold, self, jobDurationyield release, self, repairman

def timePerPart():return random.normalvariate(processingTimeMean, processingTimeSigma)

def timeToFailure():return random.expovariate(mean)

# Experiment data -------------------------

repairtime = 30.0 # minutesprocessingTimeMean = 10.0 # minutesprocessingTimeSigma = 2.0timeToFailureMean = 300.0 # minutesmean = 1 / timeToFailureMean # per minutejobDuration = 30 # minutesnrMachines = 10random.seed(111333555)weeks = 4 # weekssimTime = weeks * 24 * 60 * 7 # minutes


print('Machineshop')initialize()repairman = Resource(capacity=1, qType=PriorityQ)m = {}for i in range(nrMachines):

m[i + 1] = Machine(name="Machine {0}".format(i + 1))activate(m[i + 1], m[i + 1].working())

oj = OtherJobs()activate(oj, oj.doingJobs())



simulate(until=simTime) # minutes

# Analysis/output -------------------------

print("Machineshop results after {0} weeks".format(weeks))for i in range(nrMachines):

print("Machine {0}: {1}".format(i + 1, m[i + 1].partsMade))

MachineshopMachineshop results after 4 weeksMachine 1: 3262Machine 2: 3353Machine 3: 3161Machine 4: 3200Machine 5: 3296Machine 6: 3139Machine 7: 3287Machine 8: 3248Machine 9: 3273Machine 10: 3217


from SimPy.Simulation import *import random""" Machineshop_OO.pyMachineshop Model

An example showing interrupts and priority queuing.

Scenario:

A workshop has n identical machines. A stream of jobs (enough to keepthe machines busy) arrives. Each machine breaks downperiodically. Repairs are carried out by one repairman. The repairmanhas other, less important tasks to perform, too. Once he starts one ofthose, he completes it before starting with the machine repair. Theworkshop works continously. """


class Machine(Process):def __init__(self, name, sim):

Process.__init__(self, name=name, sim=sim)myBreaker = Breakdown(self, sim=sim)sim.activate(myBreaker, myBreaker.breakmachine())self.partsMade = 0

def working(self):while True:

yield hold, self, timePerPart()if self.interrupted():

# broken downparttimeleft = self.interruptLeftyield request, self, self.sim.repairman, 1yield hold, self, repairtime# repaired



yield release, self, self.sim.repairmanyield hold, self, parttimeleft# part completedself.partsMade += 1

else:# part madeself.partsMade += 1

class Breakdown(Process):def __init__(self, myMachine, sim):

Process.__init__(self, sim=sim)self.myMachine = myMachine

def breakmachine(self):while True:

yield hold, self, timeToFailure()self.interrupt(self.myMachine)

class OtherJobs(Process):

def doingJobs(self):while True:

yield request, self, self.sim.repairman, 0# starts working on jobsyield hold, self, jobDurationyield release, self, self.sim.repairman

def timePerPart():return random.normalvariate(processingTimeMean, processingTimeSigma)

def timeToFailure():return random.expovariate(mean)

# Experiment data -------------------------

repairtime = 30.0 # minutesprocessingTimeMean = 10.0 # minutesprocessingTimeSigma = 2.0timeToFailureMean = 300.0 # minutesmean = 1 / timeToFailureMean # per minutejobDuration = 30 # minutesnrMachines = 10random.seed(111333555)weeks = 4 # weekssimTime = weeks * 24 * 60 * 7 # minutes

# Model

class MachineshopModel(Simulation):def run(self):

print('Machineshop')self.initialize()



self.repairman = Resource(capacity=1, qType=PriorityQ, sim=self)self.m = {}for i in range(nrMachines):

self.m[i + 1] = Machine(name="Machine {0}".format(i + 1), sim=self)self.activate(self.m[i + 1], self.m[i + 1].working())

oj = OtherJobs(sim=self)self.activate(oj, oj.doingJobs())self.simulate(until=simTime) # minutes

# Experiment ------------------------------model = MachineshopModel()model.run()

# Analysis/output -------------------------

print("Machineshop results after {0} weeks".format(weeks))for i in range(nrMachines):

print("Machine {0}: {1}".format(i + 1, model.m[i + 1].partsMade))

MachineshopMachineshop results after 4 weeksMachine 1: 3262Machine 2: 3353Machine 3: 3161Machine 4: 3200Machine 5: 3296Machine 6: 3139Machine 7: 3287Machine 8: 3248Machine 9: 3273Machine 10: 3217

Supermarket: Market.py, Market_OO.py

A supermarket checkout with multiple counters and extended Monitor objects. Written and analysed by David Mertzin an article for developerWorks (). (MM)

""" Market.pyModel of a supermarket."""from SimPy.Simulation import *import randomfrom math import sqrt


class Customer(Process):def __init__(self):

Process.__init__(self)# Randomly pick how many items this customer is buyingself.items = 1 + int(random.expovariate(1.0 / AVGITEMS))

def checkout(self):start = now() # Customer decides to check out



yield request, self, checkout_aisleat_checkout = now() # Customer gets to front of linewaittime.tally(at_checkout - start)yield hold, self, self.items * ITEMTIMEleaving = now() # Customer completes purchasecheckouttime.tally(leaving - at_checkout)yield release, self, checkout_aisle

class Customer_Factory(Process):def run(self):

while 1:c = Customer()activate(c, c.checkout())arrival = random.expovariate(float(AVGCUST) / CLOSING)yield hold, self, arrival

class Monitor2(Monitor):def __init__(self):

Monitor.__init__(self)self.min, self.max = (sys.maxsize, 0)

def tally(self, x):self.observe(x)self.min = min(self.min, x)self.max = max(self.max, x)

# Experiment data -------------------------

AISLES = 6 # Number of open aislesITEMTIME = 0.1 # Time to ring up one itemAVGITEMS = 20 # Average number of items purchasedCLOSING = 60 * 12 # Minutes from store open to store closeAVGCUST = 1500 # Average number of daily customersRUNS = 8 # Number of times to run the simulationSEED = 111333555 # seed value for random numbers


random.seed(SEED)print('Market')for run in range(RUNS):

waittime = Monitor2()checkouttime = Monitor2()checkout_aisle = Resource(AISLES)initialize()cf = Customer_Factory()activate(cf, cf.run(), 0.0)simulate(until=CLOSING)FMT = "Waiting time average: {0:.1f} (std dev {1:.1f}, maximum {2:.1f})"print(FMT.format(waittime.mean(), sqrt(waittime.var()), waittime.max))

# Analysis/output -------------------------

print('AISLES:', AISLES, ' ITEM TIME:', ITEMTIME)



MarketWaiting time average: 0.5 (std dev 1.0, maximum 6.2)Waiting time average: 0.3 (std dev 0.7, maximum 3.8)Waiting time average: 0.3 (std dev 0.5, maximum 3.1)Waiting time average: 0.3 (std dev 0.7, maximum 4.1)Waiting time average: 0.3 (std dev 0.6, maximum 3.5)Waiting time average: 0.3 (std dev 0.6, maximum 4.3)Waiting time average: 0.4 (std dev 0.9, maximum 5.6)Waiting time average: 0.2 (std dev 0.5, maximum 3.7)AISLES: 6 ITEM TIME: 0.1


""" Market_OO.pyModel of a supermarket."""from SimPy.Simulation import *import randomfrom math import sqrt


class Customer(Process):def __init__(self, sim):

Process.__init__(self, sim=sim)# Randomly pick how many items this customer is buyingself.items = 1 + int(random.expovariate(1.0 / AVGITEMS))

def checkout(self):start = self.sim.now() # Customer decides to check outyield request, self, self.sim.checkout_aisleat_checkout = self.sim.now() # Customer gets to front of lineself.sim.waittime.tally(at_checkout - start)yield hold, self, self.items * ITEMTIMEleaving = self.sim.now() # Customer completes purchaseself.sim.checkouttime.tally(leaving - at_checkout)yield release, self, self.sim.checkout_aisle

class Customer_Factory(Process):def run(self):

while 1:c = Customer(sim=self.sim)self.sim.activate(c, c.checkout())arrival = random.expovariate(float(AVGCUST) / CLOSING)yield hold, self, arrival

class Monitor2(Monitor):def __init__(self, sim):

Monitor.__init__(self, sim=sim)self.min, self.max = (sys.maxsize, 0)

def tally(self, x):self.observe(x)self.min = min(self.min, x)self.max = max(self.max, x)



# Experiment data -------------------------

AISLES = 6 # Number of open aislesITEMTIME = 0.1 # Time to ring up one itemAVGITEMS = 20 # Average number of items purchasedCLOSING = 60 * 12 # Minutes from store open to store closeAVGCUST = 1500 # Average number of daily customersRUNS = 8 # Number of times to run the simulationSEED = 111333555 # seed value for random numbers

# Modelclass MarketModel(Simulation):

def runs(self):

random.seed(SEED)print('Market')for run in range(RUNS):

self.initialize()self.waittime = Monitor2(sim=self)self.checkouttime = Monitor2(sim=self)self.checkout_aisle = Resource(capacity=AISLES, sim=self)

cf = Customer_Factory(sim=self)self.activate(cf, cf.run(), 0.0)self.simulate(until=CLOSING)# Analysis/output -------------FMT = ("Waiting time average: {0:.1f} "

"(std dev {1:.1f}, maximum {2:.1f})")print(FMT.format(self.waittime.mean(),

sqrt(self.waittime.var()),self.waittime.max))

# Experiment ------------------------------MarketModel().runs()print('AISLES:', AISLES, ' ITEM TIME:', ITEMTIME)

MarketWaiting time average: 0.5 (std dev 1.0, maximum 6.2)Waiting time average: 0.3 (std dev 0.7, maximum 3.8)Waiting time average: 0.3 (std dev 0.5, maximum 3.1)Waiting time average: 0.3 (std dev 0.7, maximum 4.1)Waiting time average: 0.3 (std dev 0.6, maximum 3.5)Waiting time average: 0.3 (std dev 0.6, maximum 4.3)Waiting time average: 0.4 (std dev 0.9, maximum 5.6)Waiting time average: 0.2 (std dev 0.5, maximum 3.7)AISLES: 6 ITEM TIME: 0.1

Movie Theatre Ticket Counter: Movie_renege.py, Movie_renege_OO.py

Use of reneging (compound yield request) constructs for reneging at occurrence of an event. Scenario is a movieticket counter with a limited number of tickets for three movies (next show only). When a movie is sold out, all peoplewaiting to buy ticket for that movie renege (leave queue).



"""Movie_renege

Demo program to show event-based reneging by'yield (request,self,res),(waitevent,self,evt)'.

Scenario:A movie theatre has one ticket counter selling tickets forthree movies (next show only). When a movie is sold out, allpeople waiting to buy ticket for that movie renege (leave queue)."""from SimPy.Simulation import *from random import *


class MovieGoer(Process):def getTickets(self, whichMovie, nrTickets):

yield ((request, self, ticketCounter),(waitevent, self, soldOut[whichMovie]))

if self.acquired(ticketCounter):if available[whichMovie] >= nrTickets:

available[whichMovie] -= nrTicketsif available[whichMovie] < 2:

soldOut[whichMovie].signal()whenSoldOut[whichMovie] = now()available[whichMovie] = 0

yield hold, self, 1else:

yield hold, self, 0.5yield release, self, ticketCounter

else:nrRenegers[whichMovie] += 1

class CustomerArrivals(Process):def traffic(self):

while now() < 120:yield hold, self, expovariate(1 / 0.5)movieChoice = choice(movies)nrTickets = randint(1, 6)if available[movieChoice]:

m = MovieGoer()activate(m, m.getTickets(

whichMovie=movieChoice, nrTickets=nrTickets))

# Experiment data -------------------------seed(111333555)movies = ["Gone with the Windows", "Hard Core Dump", "Modern CPU Times"]

available = {}soldOut = {}nrRenegers = {}whenSoldOut = {}for film in movies:

available[film] = 50



soldOut[film] = SimEvent(film)nrRenegers[film] = 0

ticketCounter = Resource(capacity=1)

# Model/Experiment ------------------------------print('Movie_renege')initialize()c = CustomerArrivals()activate(c, c.traffic())simulate(until=120)

for f in movies:if soldOut[f]:

FMT = ("Movie '{0}' sold out {1:.0f} minutes after ticket counter ""opening.")

print(FMT.format(f, int(whenSoldOut[f])))FMT = "\tNr people leaving queue when film '{0}' sold out: {1}"print(FMT.format(f, nrRenegers[f]))

Movie_renegeMovie 'Gone with the Windows' sold out 41 minutes after ticket counter opening.

Nr people leaving queue when film 'Gone with the Windows' sold out: 9Movie 'Hard Core Dump' sold out 37 minutes after ticket counter opening.

Nr people leaving queue when film 'Hard Core Dump' sold out: 8Movie 'Modern CPU Times' sold out 33 minutes after ticket counter opening.

Nr people leaving queue when film 'Modern CPU Times' sold out: 9


"""Movie_renege.py

Demo program to show event-based reneging by'yield (request,self,res),(waitevent,self,evt)'.

Scenario:A movie theatre has one ticket counter selling tickets forthree movies (next show only). When a movie is sold out, allpeople waiting to buy ticket for that movie renege (leave queue)."""from SimPy.Simulation import *from random import *


class MovieGoer(Process):def getTickets(self, whichMovie, nrTickets):

yield ((request, self, self.sim.ticketCounter),(waitevent, self, self.sim.soldOut[whichMovie]))

if self.acquired(self.sim.ticketCounter):if self.sim.available[whichMovie] >= nrTickets:

self.sim.available[whichMovie] -= nrTicketsif self.sim.available[whichMovie] < 2:

self.sim.soldOut[whichMovie].signal()self.sim.whenSoldOut[whichMovie] = self.sim.now()self.sim.available[whichMovie] = 0

yield hold, self, 1else:



yield hold, self, 0.5yield release, self, self.sim.ticketCounter

else:self.sim.nrRenegers[whichMovie] += 1

class CustomerArrivals(Process):def traffic(self):

while self.sim.now() < 120:yield hold, self, expovariate(1 / 0.5)movieChoice = choice(movies)nrTickets = randint(1, 6)if self.sim.available[movieChoice]:

m = MovieGoer(sim=self.sim)self.sim.activate(m, m.getTickets(

whichMovie=movieChoice, nrTickets=nrTickets))

# Experiment data -------------------------seed(111333555)movies = ["Gone with the Windows", "Hard Core Dump", "Modern CPU Times"]

# Model -----------------------------------

class MovieRenegeModel(Simulation):def run(self):

print('Movie_renege')self.initialize()self.available = {}self.soldOut = {}self.nrRenegers = {}self.whenSoldOut = {}for film in movies:

self.available[film] = 50self.soldOut[film] = SimEvent(film, sim=self)self.nrRenegers[film] = 0

self.ticketCounter = Resource(capacity=1, sim=self)c = CustomerArrivals(sim=self)self.activate(c, c.traffic())self.simulate(until=120)

# Experiment ------------------------------model = MovieRenegeModel()model.run()

# Analysis/output -------------------------for f in movies:

if model.soldOut[f]:FMT = ("Movie '{0}' sold out {1:.0f} minutes after ticket counter "

"opening.")print(FMT.format(f, int(model.whenSoldOut[f])))FMT = "\tNr people leaving queue when film '{0}' sold out: {1}"print(FMT.format(f, model.nrRenegers[f]))

Movie_renegeMovie 'Gone with the Windows' sold out 41 minutes after ticket counter opening.



Nr people leaving queue when film 'Gone with the Windows' sold out: 9Movie 'Hard Core Dump' sold out 37 minutes after ticket counter opening.

Nr people leaving queue when film 'Hard Core Dump' sold out: 8Movie 'Modern CPU Times' sold out 33 minutes after ticket counter opening.

Nr people leaving queue when film 'Modern CPU Times' sold out: 9

Workers Sharing Tools, waitUntil: needResources.py, needResources_OO.py

Demo of waitUntil capability. It simulates three workers each requiring a set of tools to do their jobs. Tools areshared, scarce resources for which they compete.

from SimPy.Simulation import *import random

"""needResources.py

Demo of waitUntil capability.

Scenario:Three workers require sets of tools to do their jobs. Tools are shared, scarceresources for which they compete."""

class Worker(Process):def work(self, heNeeds=[]):

def workerNeeds():for item in heNeeds:

if item.n == 0:return False

return True

while now() < 8 * 60:yield waituntil, self, workerNeedsfor item in heNeeds:

yield request, self, itemprint("{0} {1} has {2} and starts job"

.format(now(),self.name,[x.name for x in heNeeds]))

yield hold, self, random.uniform(10, 30)for item in heNeeds:

yield release, self, itemyield hold, self, 2 # rest

random.seed(111333555)print('needResources')initialize()brush = Resource(capacity=1, name="brush")ladder = Resource(capacity=2, name="ladder")hammer = Resource(capacity=1, name="hammer")saw = Resource(capacity=1, name="saw")painter = Worker("painter")activate(painter, painter.work([brush, ladder]))



roofer = Worker("roofer")activate(roofer, roofer.work([hammer, ladder, ladder]))treeguy = Worker("treeguy")activate(treeguy, treeguy.work([saw, ladder]))print(simulate(until=9 * 60))

needResources0 painter has ['brush', 'ladder'] and starts job16.4985835442738 roofer has ['hammer', 'ladder', 'ladder'] and starts job39.41544751014284 treeguy has ['saw', 'ladder'] and starts job39.41544751014284 painter has ['brush', 'ladder'] and starts job56.801814464340836 roofer has ['hammer', 'ladder', 'ladder'] and starts job75.30359743269759 painter has ['brush', 'ladder'] and starts job75.30359743269759 treeguy has ['saw', 'ladder'] and starts job103.081870230719 roofer has ['hammer', 'ladder', 'ladder'] and starts job118.17856121225341 painter has ['brush', 'ladder'] and starts job118.17856121225341 treeguy has ['saw', 'ladder'] and starts job143.25197748702192 roofer has ['hammer', 'ladder', 'ladder'] and starts job170.86533741129648 treeguy has ['saw', 'ladder'] and starts job170.86533741129648 painter has ['brush', 'ladder'] and starts job196.3748629530008 roofer has ['hammer', 'ladder', 'ladder'] and starts job217.8426181665415 treeguy has ['saw', 'ladder'] and starts job217.8426181665415 painter has ['brush', 'ladder'] and starts job244.3699827271992 roofer has ['hammer', 'ladder', 'ladder'] and starts job269.8696310798061 painter has ['brush', 'ladder'] and starts job269.8696310798061 treeguy has ['saw', 'ladder'] and starts job298.28127810116723 roofer has ['hammer', 'ladder', 'ladder'] and starts job316.83616616186885 treeguy has ['saw', 'ladder'] and starts job316.83616616186885 painter has ['brush', 'ladder'] and starts job345.4248822185373 roofer has ['hammer', 'ladder', 'ladder'] and starts job359.2558219106454 painter has ['brush', 'ladder'] and starts job359.2558219106454 treeguy has ['saw', 'ladder'] and starts job385.7096335445383 roofer has ['hammer', 'ladder', 'ladder'] and starts job407.43697134395245 painter has ['brush', 'ladder'] and starts job407.43697134395245 treeguy has ['saw', 'ladder'] and starts job427.5649730427561 roofer has ['hammer', 'ladder', 'ladder'] and starts job451.6157857370266 painter has ['brush', 'ladder'] and starts job451.6157857370266 treeguy has ['saw', 'ladder'] and starts job472.97306394134876 roofer has ['hammer', 'ladder', 'ladder'] and starts job492.3976188422561 treeguy has ['saw', 'ladder'] and starts job492.3976188422561 painter has ['brush', 'ladder'] and starts jobSimPy: No more events at time 515.0226394595458


from SimPy.Simulation import *import random

"""needResources.py

Demo of waitUntil capability.

Scenario:Three workers require sets of tools to do their jobs. Tools are shared, scarceresources for which they compete."""




class Worker(Process):def work(self, heNeeds=[]):

def workerNeeds():for item in heNeeds:

if item.n == 0:return False

return True

while self.sim.now() < 8 * 60:yield waituntil, self, workerNeedsfor item in heNeeds:

yield request, self, itemprint("{0} {1} has {2} and starts job"

.format(self.sim.now(),self.name,[x.name for x in heNeeds]))

yield hold, self, random.uniform(10, 30)for item in heNeeds:

yield release, self, itemyield hold, self, 2 # rest

# Model -----------------------------------class NeedResourcesModel(Simulation):

def run(self):print('needResources')self.initialize()brush = Resource(capacity=1, name="brush", sim=self)ladder = Resource(capacity=2, name="ladder", sim=self)hammer = Resource(capacity=1, name="hammer", sim=self)saw = Resource(capacity=1, name="saw", sim=self)painter = Worker("painter", sim=self)self.activate(painter, painter.work([brush, ladder]))roofer = Worker("roofer", sim=self)self.activate(roofer, roofer.work([hammer, ladder, ladder]))treeguy = Worker("treeguy", sim=self)self.activate(treeguy, treeguy.work([saw, ladder]))print(self.simulate(until=9 * 60))

# Experiment data -------------------------SEED = 111333555

# Experiment ------------------------------random.seed(SEED)NeedResourcesModel().run()

needResources0 painter has ['brush', 'ladder'] and starts job16.4985835442738 roofer has ['hammer', 'ladder', 'ladder'] and starts job39.41544751014284 treeguy has ['saw', 'ladder'] and starts job39.41544751014284 painter has ['brush', 'ladder'] and starts job56.801814464340836 roofer has ['hammer', 'ladder', 'ladder'] and starts job75.30359743269759 painter has ['brush', 'ladder'] and starts job75.30359743269759 treeguy has ['saw', 'ladder'] and starts job



103.081870230719 roofer has ['hammer', 'ladder', 'ladder'] and starts job118.17856121225341 painter has ['brush', 'ladder'] and starts job118.17856121225341 treeguy has ['saw', 'ladder'] and starts job143.25197748702192 roofer has ['hammer', 'ladder', 'ladder'] and starts job170.86533741129648 treeguy has ['saw', 'ladder'] and starts job170.86533741129648 painter has ['brush', 'ladder'] and starts job196.3748629530008 roofer has ['hammer', 'ladder', 'ladder'] and starts job217.8426181665415 treeguy has ['saw', 'ladder'] and starts job217.8426181665415 painter has ['brush', 'ladder'] and starts job244.3699827271992 roofer has ['hammer', 'ladder', 'ladder'] and starts job269.8696310798061 painter has ['brush', 'ladder'] and starts job269.8696310798061 treeguy has ['saw', 'ladder'] and starts job298.28127810116723 roofer has ['hammer', 'ladder', 'ladder'] and starts job316.83616616186885 treeguy has ['saw', 'ladder'] and starts job316.83616616186885 painter has ['brush', 'ladder'] and starts job345.4248822185373 roofer has ['hammer', 'ladder', 'ladder'] and starts job359.2558219106454 painter has ['brush', 'ladder'] and starts job359.2558219106454 treeguy has ['saw', 'ladder'] and starts job385.7096335445383 roofer has ['hammer', 'ladder', 'ladder'] and starts job407.43697134395245 painter has ['brush', 'ladder'] and starts job407.43697134395245 treeguy has ['saw', 'ladder'] and starts job427.5649730427561 roofer has ['hammer', 'ladder', 'ladder'] and starts job451.6157857370266 painter has ['brush', 'ladder'] and starts job451.6157857370266 treeguy has ['saw', 'ladder'] and starts job472.97306394134876 roofer has ['hammer', 'ladder', 'ladder'] and starts job492.3976188422561 treeguy has ['saw', 'ladder'] and starts job492.3976188422561 painter has ['brush', 'ladder'] and starts jobSimPy: No more events at time 515.0226394595458

Widget Factory: SimPy_worker_extend.py, SimPy_worker_extend_OO.py

Factory making widgets with queues for machines. (MM)

from SimPy.Simulation import *from random import uniform, seed

def theTime(time):

hrs = int(time / 60)min = int(time - hrs * 60)return "{0}:{1}".format(str.zfill(str(hrs), 2), str.zfill(str(min), 2))

class worker(Process):def __init__(self, id):

Process.__init__(self)self.id = idself.output = 0self.idle = 0self.total_idle = 0

def working_day(self, foobar):print("{0} Worker {1} arrives in factory".format(

theTime(now()), self.id))while now() < 17 * 60: # work till 5 pm



yield hold, self, uniform(3, 10)# print("{0} Widget completed".format(theTime(now())))foobar.queue.append(self)if foobar.idle:

reactivate(foobar)else:

self.idle = 1 # worker has to wait for foobar servicestart_idle = now()# print("{0} Worker {1} queuing for foobar machine"# .format(theTime(now()),self.id))

yield passivate, self # waiting and foobar serviceself.output += 1if self.idle:

self.total_idle += now() - start_idleself.idle = 0

print("{0} {1} goes home, having built {2:d} widgets today.".format(theTime(now()), self.id, self.output))

print("Worker {0} was idle waiting for foobar machine for ""{1:3.1f} hours".format(self.id, self.total_idle / 60))

class foobar_machine(Process):def __init__(self):

Process.__init__(self)self.queue = []self.idle = 1

def foobar_Process(self):yield passivate, selfwhile 1:

while len(self.queue) > 0:self.idle = 0yield hold, self, 3served = self.queue.pop(0)reactivate(served)

self.idle = 1yield passivate, self

seed(111333555)print('SimPy_worker_extend')initialize()foo = foobar_machine()activate(foo, foo.foobar_Process(), delay=0)john = worker("John")activate(john, john.working_day(foo), at=510) # start at 8:30 ameve = worker("Eve")activate(eve, eve.working_day(foo), at=510)simulate(until=18 * 60)# scheduler(till=18*60) #run simulation from midnight till 6 pm

SimPy_worker_extend08:30 Worker John arrives in factory08:30 Worker Eve arrives in factory17:03 Eve goes home, having built 52 widgets today.Worker Eve was idle waiting for foobar machine for 1.3 hours17:09 John goes home, having built 52 widgets today.Worker John was idle waiting for foobar machine for 0.8 hours




from SimPy.Simulation import *from random import uniform, seed


def theTime(time):

hrs = int(time / 60)min = int(time - hrs * 60)return "{0}:{1}".format(str.zfill(str(hrs), 2), str.zfill(str(min), 2))

class worker(Process):def __init__(self, id, sim):

Process.__init__(self, sim=sim)self.id = idself.output = 0self.idle = 0self.total_idle = 0

def working_day(self, foobar):print("{0} Worker {1} arrives in factory".format(

theTime(self.sim.now()), self.id))while self.sim.now() < 17 * 60: # work till 5 pm

yield hold, self, uniform(3, 10)# print("{0} Widget completed".format(theTime(now())))foobar.queue.append(self)if foobar.idle:

self.sim.reactivate(foobar)else:

self.idle = 1 # worker has to wait for foobar servicestart_idle = self.sim.now()# print("{0} Worker {1} queuing for foobar machine"# .format(theTime(now()),self.id))

yield passivate, self # waiting and foobar serviceself.output += 1if self.idle:

self.total_idle += self.sim.now() - start_idleself.idle = 0

print("{0} {1} goes home, having built {2} widgets today.".format(theTime(self.sim.now()), self.id, self.output))

print("Worker {0} was idle waiting for foobar machine for ""{1:3.1f} hours".format(self.id, self.total_idle / 60))

class foobar_machine(Process):def __init__(self, sim):

Process.__init__(self, sim=sim)self.queue = []self.idle = 1

def foobar_Process(self):yield passivate, selfwhile 1:



while len(self.queue) > 0:self.idle = 0yield hold, self, 3served = self.queue.pop(0)self.sim.reactivate(served)

self.idle = 1yield passivate, self

# Model -----------------------------------

class SimPy_Worker_Extend_Model(Simulation):def run(self):

print('SimPy_worker_extend')self.initialize()foo = foobar_machine(sim=self)self.activate(foo, foo.foobar_Process(), delay=0)john = worker("John", sim=self)self.activate(john, john.working_day(foo), at=510) # start at 8:30 ameve = worker("Eve", sim=self)self.activate(eve, eve.working_day(foo), at=510)self.simulate(until=18 * 60) # run simulation from midnight till 6 pm

# Expriment -------------------------------seed(111333555)SimPy_Worker_Extend_Model().run()

SimPy_worker_extend08:30 Worker John arrives in factory08:30 Worker Eve arrives in factory17:03 Eve goes home, having built 52 widgets today.Worker Eve was idle waiting for foobar machine for 1.3 hours17:09 John goes home, having built 52 widgets today.Worker John was idle waiting for foobar machine for 0.8 hours

Widget Packing Machine: WidgetPacking.py, WidgetPacking_OO.py

Using buffers for producer/consumer scenarios. Scenario is a group of widget producing machines and a widgetpacker, synchronised via a buffer. Two models are shown: the first uses a Level for buffering non-distinguishableitems (widgets), and the second a Store for distinguishable items (widgets of different weight).

"""WidgetPacking.pyScenario:In a factory, nProd widget-making machines produce widgets with a weightwidgWeight (uniformly distributed [widgWeight-dWeight..widgWeight+dWeight])on average every tMake minutes (uniform distribution[tmake-deltaT..tMake+deltaT]).A widget-packing machine packs widgets in tPackminutes into packages. Simulate for simTime minutes.

Model 1:The widget-packer packs nWidgets into a package.

Model 2:The widget-packer packs widgets into a package with package weight notto exceed packMax.



"""from SimPy.Simulation import *import random

# Experiment data -------------------------nProd = 2 # widget-making machineswidgWeight = 20 # kilogrammesdWeight = 3 # kilogrammestMake = 2 # minutesdeltaT = 0.75 # minutestPack = 0.25 # minutes per idgetinitialSeed = 1234567 # for random number streamsimTime = 100 # minutes

print('WidgetPacking')################################################################## Model 1################################################################### DatanWidgets = 6 # widgets per package

class Widget:def __init__(self, weight):


class WidgetMakerN(Process):"""Produces widgets"""

def make(self, buffer):while True:

yield hold, self, r.uniform(tMake - deltaT, tMake + deltaT)yield put, self, buffer, 1 # buffer 1 widget

class WidgetPackerN(Process):"""Packs a number of widgets into a package"""

def pack(self, buffer):while True:

for i in range(nWidgets):yield get, self, buffer, 1 # get widgetyield hold, self, tPack # pack it

print("{0}: package completed".format(now()))

print("Model 1: pack {0} widgets per package".format(nWidgets))initialize()r = random.Random()r.seed(initialSeed)wBuffer = Level(name="WidgetBuffer", capacity=500)for i in range(nProd):

wm = WidgetMakerN(name="WidgetMaker{0}".format(i))activate(wm, wm.make(wBuffer))

wp = WidgetPackerN(name="WidgetPacker")



activate(wp, wp.pack(wBuffer))simulate(until=simTime)

################################################################## Model 2################################################################### DatapackMax = 120 # kilogrammes per package (max)

class WidgetMakerW(Process):"""Produces widgets"""


yield hold, self, r.uniform(tMake - deltaT, tMake + deltaT)widgetWeight = r.uniform(

widgWeight - dWeight, widgWeight + dWeight)# buffer widgetyield put, self, buffer, [Widget(weight=widgetWeight)]

class WidgetPackerW(Process):"""Packs a number of widgets into a package"""

def pack(self, buffer):weightLeft = 0while True:

packWeight = weightLeft # pack a widget which did not fit# into previous packageweightLeft = 0while True:

yield get, self, buffer, 1 # get widgetweightReceived = self.got[0].weightif weightReceived + packWeight <= packMax:

yield hold, self, tPack # pack itpackWeight += weightReceived

else:weightLeft = weightReceived # for next packagebreak

print("{0}: package completed. Weight= {1:6.2f} kg".format(now(), packWeight))

print("\nModel 2: pack widgets up to max package weight of {0}".format(packMax))

initialize()r = random.Random()r.seed(initialSeed)wBuffer = Store(name="WidgetBuffer", capacity=500)for i in range(nProd):

wm = WidgetMakerW(name="WidgetMaker{0}".format(i))activate(wm, wm.make(wBuffer))

wp = WidgetPackerW(name="WidgetPacker")activate(wp, wp.pack(wBuffer))simulate(until=simTime)



WidgetPackingModel 1: pack 6 widgets per package6.925710569651607: package completed12.932859800562765: package completed19.296609507200795: package completed25.827518016011588: package completed32.28961167417064: package completed38.27653780313223: package completed43.516005487393166: package completed49.84136801349117: package completed56.02183282676389: package completed61.22427700160065: package completed66.78084592723073: package completed73.45267491487972: package completed78.52559623726368: package completed84.36238918070774: package completed91.00334807967428: package completed96.25112005144973: package completed

Model 2: pack widgets up to max package weight of 1201.8535993255800536: package completed. Weight= 20.49 kg2.9447148714144333: package completed. Weight= 43.29 kg3.443679099583584: package completed. Weight= 60.73 kg5.111979202993875: package completed. Weight= 81.36 kg5.495036360384358: package completed. Weight= 103.32 kg7.53958534514355: package completed. Weight= 43.79 kg9.454524267565933: package completed. Weight= 62.32 kg9.704524267565933: package completed. Weight= 82.07 kg11.223206345291937: package completed. Weight= 103.55 kg13.534262617168128: package completed. Weight= 41.81 kg13.930945836966112: package completed. Weight= 62.71 kg15.109021173114078: package completed. Weight= 82.73 kg15.9312940569179: package completed. Weight= 103.41 kg18.483473187078456: package completed. Weight= 37.07 kg19.489762737880447: package completed. Weight= 56.79 kg20.019035270021426: package completed. Weight= 77.23 kg21.516971139857024: package completed. Weight= 96.93 kg21.922330284736397: package completed. Weight= 116.85 kg24.14512998963336: package completed. Weight= 37.89 kg24.39877824367329: package completed. Weight= 55.50 kg26.67484997458752: package completed. Weight= 78.46 kg26.92484997458752: package completed. Weight= 100.35 kg28.1664758950799: package completed. Weight= 117.54 kg29.69269527817336: package completed. Weight= 40.05 kg31.039995345892564: package completed. Weight= 60.53 kg31.416541860477725: package completed. Weight= 77.63 kg32.41760810512986: package completed. Weight= 100.53 kg32.95461868723966: package completed. Weight= 118.77 kg35.03241015535913: package completed. Weight= 40.09 kg37.13500099371781: package completed. Weight= 59.57 kg37.54993412593156: package completed. Weight= 76.58 kg39.30652910593313: package completed. Weight= 95.59 kg39.55652910593313: package completed. Weight= 116.49 kg42.0420842938092: package completed. Weight= 42.24 kg42.2920842938092: package completed. Weight= 63.58 kg43.65741443896232: package completed. Weight= 85.90 kg43.90741443896232: package completed. Weight= 106.78 kg46.0951882462178: package completed. Weight= 37.20 kg



47.438300452780354: package completed. Weight= 58.47 kg47.688300452780354: package completed. Weight= 79.33 kg49.40690976731008: package completed. Weight= 100.51 kg50.09589376227701: package completed. Weight= 117.80 kg52.015927666123616: package completed. Weight= 37.47 kg53.308294328826435: package completed. Weight= 58.78 kg54.03538112053091: package completed. Weight= 81.10 kg55.13265226734849: package completed. Weight= 98.94 kg56.54061780546726: package completed. Weight= 39.67 kg58.42284478038792: package completed. Weight= 61.78 kg58.67284478038792: package completed. Weight= 80.36 kg60.363861577200076: package completed. Weight= 102.50 kg62.30054431149251: package completed. Weight= 38.39 kg62.72667501529395: package completed. Weight= 58.21 kg64.04176585171321: package completed. Weight= 75.31 kg64.78871826102055: package completed. Weight= 94.49 kg65.5200550350699: package completed. Weight= 113.54 kg67.70604486309713: package completed. Weight= 41.53 kg69.2327615315688: package completed. Weight= 60.57 kg69.64556242167215: package completed. Weight= 82.52 kg70.56237737140228: package completed. Weight= 104.26 kg72.1477929466095: package completed. Weight= 41.62 kg72.98432452690808: package completed. Weight= 63.88 kg73.5826902990221: package completed. Weight= 81.87 kg75.05025424304776: package completed. Weight= 101.04 kg77.26048662562239: package completed. Weight= 40.26 kg77.71710036932708: package completed. Weight= 61.69 kg79.75353580124: package completed. Weight= 80.72 kg80.12811805319318: package completed. Weight= 100.29 kg82.4142311699317: package completed. Weight= 39.35 kg83.68305885096552: package completed. Weight= 57.13 kg83.93305885096552: package completed. Weight= 74.55 kg85.85481524806448: package completed. Weight= 93.29 kg86.13093377784419: package completed. Weight= 115.05 kg87.86043544925184: package completed. Weight= 41.13 kg89.46190021907331: package completed. Weight= 63.30 kg90.40626417525424: package completed. Weight= 82.59 kg91.19235409838856: package completed. Weight= 105.00 kg92.48486291641056: package completed. Weight= 37.14 kg93.79148753229941: package completed. Weight= 56.83 kg94.19510234363693: package completed. Weight= 76.57 kg96.09493981469757: package completed. Weight= 94.85 kg96.82617211079959: package completed. Weight= 114.58 kg98.82720299189798: package completed. Weight= 39.12 kg99.94853712506526: package completed. Weight= 58.33 kg


"""WidgetPacking_OO.pyScenario:In a factory, nProd widget-making machines produce widgets with a weightwidgWeight (uniformly distributed [widgWeight-dWeight..widgWeight+dWeight])on average every tMake minutes (uniform distribution[tmake-deltaT..tMake+deltaT]).A widget-packing machine packs widgets in tPackminutes into packages. Simulate for simTime minutes.

Model 1:The widget-packer packs nWidgets into a package.



Model 2:The widget-packer packs widgets into a package with package weight notto exceed packMax."""from SimPy.Simulation import *import random

# Experiment data -------------------------nProd = 2 # widget-making machineswidgWeight = 20 # kilogrammesdWeight = 3 # kilogrammestMake = 2 # minutesdeltaT = 0.75 # minutestPack = 0.25 # minutes per idgetinitialSeed = 1234567 # for random number streamsimTime = 100 # minutes

print('WidgetPacking')


# DatanWidgets = 6 # widgets per package

class Widget:def __init__(self, weight):


class WidgetMakerN(Process):def make(self, buffer):

while True:yield hold, self, self.sim.r.uniform(tMake - deltaT,

tMake + deltaT)yield put, self, buffer, 1 # buffer 1 widget

class WidgetPackerN(Process):"""Packs a number of widgets into a package"""

def pack(self, buffer):while True:

for i in range(nWidgets):yield get, self, buffer, 1 # get widgetyield hold, self, tPack # pack it

print("{0}: package completed".format(self.sim.now()))

# Model 1 ---------------------------------

class WidgetPackingModel1(Simulation):def run(self):

print("Model 1: pack {0} widgets per package".format(nWidgets))self.initialize()self.r = random.Random()self.r.seed(initialSeed)



wBuffer = Level(name="WidgetBuffer", capacity=500, sim=self)for i in range(nProd):

wm = WidgetMakerN(name="WidgetMaker{0}".format(i), sim=self)self.activate(wm, wm.make(wBuffer))

wp = WidgetPackerN(name="WidgetPacker", sim=self)self.activate(wp, wp.pack(wBuffer))self.simulate(until=simTime)

# Experiment ------------------------------WidgetPackingModel1().run()


# DatapackMax = 120 # kilogrammes per package (max)

class WidgetMakerW(Process):"""Produces widgets"""


yield hold, self, self.sim.r.uniform(tMake - deltaT,tMake + deltaT)

widgetWeight = self.sim.r.uniform(widgWeight - dWeight, widgWeight + dWeight)

# buffer widgetyield put, self, buffer, [Widget(weight=widgetWeight)]

class WidgetPackerW(Process):"""Packs a number of widgets into a package"""

def pack(self, buffer):weightLeft = 0while True:

packWeight = weightLeft # pack a widget which did not fit# into previous packageweightLeft = 0while True:

yield get, self, buffer, 1 # get widgetweightReceived = self.got[0].weightif weightReceived + packWeight <= packMax:

yield hold, self, tPack # pack itpackWeight += weightReceived

else:weightLeft = weightReceived # for next packagebreak

print("{0}: package completed. Weight= {1:6.2f} kg".format(self.sim.now(), packWeight))

# Model 2 ---------------------------------

class WidgetPackingModel2(Simulation):def run(self):

print(



"\nModel 2: pack widgets up to max package weight of {0}".format(packMax))

self.initialize()self.r = random.Random()self.r.seed(initialSeed)wBuffer = Store(name="WidgetBuffer", capacity=500, sim=self)for i in range(nProd):

wm = WidgetMakerW(name="WidgetMaker{0}".format(i), sim=self)self.activate(wm, wm.make(wBuffer))

wp = WidgetPackerW(name="WidgetPacker", sim=self)self.activate(wp, wp.pack(wBuffer))self.simulate(until=simTime)

# Experiment ------------------------------WidgetPackingModel2().run()

WidgetPackingModel 1: pack 6 widgets per package6.925710569651607: package completed12.932859800562765: package completed19.296609507200795: package completed25.827518016011588: package completed32.28961167417064: package completed38.27653780313223: package completed43.516005487393166: package completed49.84136801349117: package completed56.02183282676389: package completed61.22427700160065: package completed66.78084592723073: package completed73.45267491487972: package completed78.52559623726368: package completed84.36238918070774: package completed91.00334807967428: package completed96.25112005144973: package completed

Model 2: pack widgets up to max package weight of 1201.8535993255800536: package completed. Weight= 20.49 kg2.9447148714144333: package completed. Weight= 43.29 kg3.443679099583584: package completed. Weight= 60.73 kg5.111979202993875: package completed. Weight= 81.36 kg5.495036360384358: package completed. Weight= 103.32 kg7.53958534514355: package completed. Weight= 43.79 kg9.454524267565933: package completed. Weight= 62.32 kg9.704524267565933: package completed. Weight= 82.07 kg11.223206345291937: package completed. Weight= 103.55 kg13.534262617168128: package completed. Weight= 41.81 kg13.930945836966112: package completed. Weight= 62.71 kg15.109021173114078: package completed. Weight= 82.73 kg15.9312940569179: package completed. Weight= 103.41 kg18.483473187078456: package completed. Weight= 37.07 kg19.489762737880447: package completed. Weight= 56.79 kg20.019035270021426: package completed. Weight= 77.23 kg21.516971139857024: package completed. Weight= 96.93 kg21.922330284736397: package completed. Weight= 116.85 kg24.14512998963336: package completed. Weight= 37.89 kg24.39877824367329: package completed. Weight= 55.50 kg26.67484997458752: package completed. Weight= 78.46 kg



26.92484997458752: package completed. Weight= 100.35 kg28.1664758950799: package completed. Weight= 117.54 kg29.69269527817336: package completed. Weight= 40.05 kg31.039995345892564: package completed. Weight= 60.53 kg31.416541860477725: package completed. Weight= 77.63 kg32.41760810512986: package completed. Weight= 100.53 kg32.95461868723966: package completed. Weight= 118.77 kg35.03241015535913: package completed. Weight= 40.09 kg37.13500099371781: package completed. Weight= 59.57 kg37.54993412593156: package completed. Weight= 76.58 kg39.30652910593313: package completed. Weight= 95.59 kg39.55652910593313: package completed. Weight= 116.49 kg42.0420842938092: package completed. Weight= 42.24 kg42.2920842938092: package completed. Weight= 63.58 kg43.65741443896232: package completed. Weight= 85.90 kg43.90741443896232: package completed. Weight= 106.78 kg46.0951882462178: package completed. Weight= 37.20 kg47.438300452780354: package completed. Weight= 58.47 kg47.688300452780354: package completed. Weight= 79.33 kg49.40690976731008: package completed. Weight= 100.51 kg50.09589376227701: package completed. Weight= 117.80 kg52.015927666123616: package completed. Weight= 37.47 kg53.308294328826435: package completed. Weight= 58.78 kg54.03538112053091: package completed. Weight= 81.10 kg55.13265226734849: package completed. Weight= 98.94 kg56.54061780546726: package completed. Weight= 39.67 kg58.42284478038792: package completed. Weight= 61.78 kg58.67284478038792: package completed. Weight= 80.36 kg60.363861577200076: package completed. Weight= 102.50 kg62.30054431149251: package completed. Weight= 38.39 kg62.72667501529395: package completed. Weight= 58.21 kg64.04176585171321: package completed. Weight= 75.31 kg64.78871826102055: package completed. Weight= 94.49 kg65.5200550350699: package completed. Weight= 113.54 kg67.70604486309713: package completed. Weight= 41.53 kg69.2327615315688: package completed. Weight= 60.57 kg69.64556242167215: package completed. Weight= 82.52 kg70.56237737140228: package completed. Weight= 104.26 kg72.1477929466095: package completed. Weight= 41.62 kg72.98432452690808: package completed. Weight= 63.88 kg73.5826902990221: package completed. Weight= 81.87 kg75.05025424304776: package completed. Weight= 101.04 kg77.26048662562239: package completed. Weight= 40.26 kg77.71710036932708: package completed. Weight= 61.69 kg79.75353580124: package completed. Weight= 80.72 kg80.12811805319318: package completed. Weight= 100.29 kg82.4142311699317: package completed. Weight= 39.35 kg83.68305885096552: package completed. Weight= 57.13 kg83.93305885096552: package completed. Weight= 74.55 kg85.85481524806448: package completed. Weight= 93.29 kg86.13093377784419: package completed. Weight= 115.05 kg87.86043544925184: package completed. Weight= 41.13 kg89.46190021907331: package completed. Weight= 63.30 kg90.40626417525424: package completed. Weight= 82.59 kg91.19235409838856: package completed. Weight= 105.00 kg92.48486291641056: package completed. Weight= 37.14 kg93.79148753229941: package completed. Weight= 56.83 kg94.19510234363693: package completed. Weight= 76.57 kg



96.09493981469757: package completed. Weight= 94.85 kg96.82617211079959: package completed. Weight= 114.58 kg98.82720299189798: package completed. Weight= 39.12 kg99.94853712506526: package completed. Weight= 58.33 kg

GUI Input

The GUI examples do not run under Python 3.x, as only the core SimPy libraries were ported.

Fireworks using SimGUI: GUIdemo.py, GUIdemo_OO.py

A firework show. This is a very basic model, demonstrating the ease of interfacing to SimGUI.

__doc__ = """ GUIdemo.pyThis is a very basic model, demonstrating the easeof interfacing to SimGUI."""from SimPy.Simulation import *from random import *from SimPy.SimGUI import *


class Launcher(Process):nrLaunched = 0


gui.writeConsole("Launch at {0:.1f}".format(now()))Launcher.nrLaunched += 1gui.launchmonitor.observe(Launcher.nrLaunched)yield hold, self, uniform(1, gui.params.maxFlightTime)gui.writeConsole("Boom!!! Aaaah!! at {0:.1f}".format(now()))

def model():gui.launchmonitor = Monitor(name="Rocket counter",

ylab="nr launched", tlab="time")initialize()Launcher.nrLaunched = 0for i in range(gui.params.nrLaunchers):

lau = Launcher()activate(lau, lau.launch())

simulate(until=gui.params.duration)gui.noRunYet = Falsegui.writeStatusLine("{0} rockets launched in {1:.1f} minutes"

.format(Launcher.nrLaunched, now()))

class MyGUI(SimGUI):def __init__(self, win, **par):

SimGUI.__init__(self, win, **par)self.run.add_command(label="Start fireworks",

command=model, underline=0)



self.params = Parameters(duration=duration,maxFlightTime=maxFlightTime,nrLaunchers=nrLaunchers)

# Experiment data ---------------------------------------

duration = 2000maxFlightTime = 11.7nrLaunchers = 3

# Model/Experiment/Display ------------------------------

root = Tk()gui = MyGUI(root, title="RocketGUI", doc=__doc__, consoleHeight=40)gui.mainloop()


__doc__ = """ GUIdemo_OO.pyThis is a very basic model, demonstrating the easeof interfacing to SimGUI."""from SimPy.Simulation import *from random import *from SimPy.SimGUI import *

# Model components ---------------------------------------

class Launcher(Process):nrLaunched = 0


gui.writeConsole("Launch at {0:.1f}".format(self.sim.now()))Launcher.nrLaunched += 1gui.launchmonitor.observe(Launcher.nrLaunched)yield hold, self, uniform(1, gui.params.maxFlightTime)gui.writeConsole(

"Boom!!! Aaaah!! at {0:.1f}".format(self.sim.now()))

# Model --------------------------------------------------

class GUIdemoModel(Simulation):def run(self):

self.initialize()gui.launchmonitor = Monitor(name="Rocket counter",

ylab="nr launched", tlab="time", sim=self)Launcher.nrLaunched = 0for i in range(gui.params.nrLaunchers):

lau = Launcher(sim=self)self.activate(lau, lau.launch())

self.simulate(until=gui.params.duration)gui.noRunYet = Falsegui.writeStatusLine("{0} rockets launched in {1:.1f} minutes"

.format(Launcher.nrLaunched, self.now()))



# Model GUI ----------------------------------------------

class MyGUI(SimGUI):def __init__(self, win, **par):

SimGUI.__init__(self, win, **par)self.run.add_command(label="Start fireworks",

command=GUIdemoModel().run, underline=0)self.params = Parameters(duration=duration,

maxFlightTime=maxFlightTime,nrLaunchers=nrLaunchers)

# Experiment data ----------------------------------------

duration = 2000maxFlightTime = 11.7nrLaunchers = 3

# Experiment/Display ------------------------------------

root = Tk()gui = MyGUI(root, title="RocketGUI", doc=__doc__, consoleHeight=40)gui.mainloop()

Bank Customers using SimGUI: bank11GUI.py, bank11GUI_OO.py

Simulation with customers arriving at random to a bank with two counters. This is a modification of the bank11simulation using SimGUI to run the simulation.

__doc__ = """ bank11.py: Simulate customers arrivingat random, using a Source, requesting servicefrom two counters each with their own queuerandom servicetime.Uses a Monitor object to record waiting times

"""from SimPy.Simulation import *# Lmonafrom random import Randomfrom SimPy.SimGUI import *


def __init__(self, seed=333):Process.__init__(self)self.SEED = seed


c = Customer(name="Customer{0:02d}".format(i))activate(c, c.visit(timeInBank=12.0))t = rv.expovariate(1.0 / interval)



yield hold, self, t

def NoInSystem(R):""" The number of customers in the resource Rin waitQ and active Q"""return (len(R.waitQ) + len(R.activeQ))


def __init__(self, name):Process.__init__(self)self.name = name

def visit(self, timeInBank=0):arrive = now()Qlength = [NoInSystem(counter[i]) for i in range(Nc)]if gui.params.trace:

gui.writeConsole("{0:7.4f} {1}: Here I am. Queues are: {2}".format(now(), self.name, Qlength))

for i in range(Nc):if Qlength[i] == 0 or Qlength[i] == min(Qlength):

join = ibreak

yield request, self, counter[join]wait = now() - arrivewaitMonitor.observe(wait, t=now())if gui.params.trace:

gui.writeConsole("{0:7.4f} {1}: Waited {2:6.3f}".format(now(), self.name, wait))

tib = counterRV.expovariate(1.0 / timeInBank)yield hold, self, tibyield release, self, counter[join]if gui.params.trace:

gui.writeConsole("{0:7.4f} {1}: Finished ".format(now(), self.name))

def model(counterseed=3939393):global Nc, counter, counterRV, waitMonitorNc = 2counter = [Resource(name="Clerk0"), Resource(name="Clerk1")]counterRV = Random(counterseed)gui.mon1 = waitMonitor = Monitor("Customer waiting times")waitMonitor.tlab = "arrival time"waitMonitor.ylab = "waiting time"initialize()source = Source(seed=gui.params.sourceseed)activate(source, source.generate(

gui.params.numberCustomers, gui.params.interval), 0.0)result = simulate(until=gui.params.endtime)gui.writeStatusLine(text="Time at simulation end: {0:.1f}"

" -- Customers still waiting: {1}".format(now(),

len(counter[0].waitQ) +len(counter[1].waitQ)))



def statistics():if gui.noRunYet:

showwarning(title='Model warning',message="Run simulation first -- no data available.")

returngui.writeConsole(text="\nRun parameters: {0}".format(gui.params))gui.writeConsole(text="Average wait for {0:4d} customers was {1:6.2f}"

.format(waitMonitor.count(), waitMonitor.mean()))

def run():model(gui.params.counterseed)gui.noRunYet = False

def showAuthors():gui.showTextBox(text="Tony Vignaux\nKlaus Muller",

title="Author information")

class MyGUI(SimGUI):def __init__(self, win, **p):

SimGUI.__init__(self, win, **p)self.help.add_command(label="Author(s)",

command=showAuthors, underline=0)self.view.add_command(label="Statistics",

command=statistics, underline=0)self.run.add_command(label="Run bank11 model",

command=run, underline=0)

self.params = Parameters(endtime=2000,sourceseed=1133,counterseed=3939393,numberCustomers=50,interval=10.0,trace=0)

root = Tk()gui = MyGUI(root, title="SimPy GUI example", doc=__doc__, consoleHeight=40)gui.mainloop()


__doc__ = """ bank11_OO.py: Simulate customers arrivingat random, using a Source, requesting servicefrom two counters each with their own queuerandom servicetime.Uses a Monitor object to record waiting times

"""from SimPy.Simulation import *from random import Randomfrom SimPy.SimGUI import *




def __init__(self, sim, seed=333):Process.__init__(self, sim=sim)self.SEED = seed


c = Customer(name="Customer{0:02d}".format(i), sim=self.sim)self.sim.activate(c, c.visit(timeInBank=12.0))t = rv.expovariate(1.0 / interval)yield hold, self, t


class Customer(Process):""" Customer arrives, is served and leaves """# def __init__(self,name,sim):# Process.__init__(self,name=name,sim=sim)

def visit(self, timeInBank=0):arrive = self.sim.now()Qlength = [NoInSystem(self.sim.counter[i]) for i in range(self.sim.Nc)]if gui.params.trace:

gui.writeConsole("{0:7.4f} {1}: Here I am. Queues are: {2}".format(self.sim.now(), self.name, Qlength))

for i in range(self.sim.Nc):if Qlength[i] == 0 or Qlength[i] == min(Qlength):

join = ibreak

yield request, self, self.sim.counter[join]wait = self.sim.now() - arriveself.sim.waitMonitor.observe(wait, t=self.sim.now())if gui.params.trace:

gui.writeConsole("{0:7.4f} {1}: Waited {2:6.3f}".format(self.sim.now(), self.name, wait))

tib = self.sim.counterRV.expovariate(1.0 / timeInBank)yield hold, self, tibyield release, self, self.sim.counter[join]if gui.params.trace:

gui.writeConsole("{0:7.4f} {1}: Finished ".format(self.sim.now(), self.name))

class CounterModel(Simulation):def run(self, counterseed=3939393):

self.initialize()self.Nc = 2self.counter = [Resource(name="Clerk0", sim=self),

Resource(name="Clerk1", sim=self)]self.counterRV = Random(counterseed)



gui.mon1 = self.waitMonitor = Monitor("Customer waiting times", sim=self)

self.waitMonitor.tlab = "arrival time"self.waitMonitor.ylab = "waiting time"source = Source(seed=gui.params.sourceseed, sim=self)self.activate(source, source.generate(

gui.params.numberCustomers, gui.params.interval), 0.0)result = self.simulate(until=gui.params.endtime)gui.writeStatusLine(text="Time at simulation end: {0:.1f}"

" -- Customers still waiting: {1}".format(self.now(),

len(self.counter[0].waitQ) +len(self.counter[1].waitQ)))


showwarning(title='Model warning',message="Run simulation first -- no data available.")

returngui.writeConsole(text="\nRun parameters: {0}".format(gui.params))gui.writeConsole(text="Average wait for {0:4d} customers was {1:6.2f}"

.format(gui.mon1.count(), gui.mon1.mean()))

def run():CounterModel().run(gui.params.counterseed)gui.noRunYet = False




SimGUI.__init__(self, win, **p)self.help.add_command(label="Author(s)",

command=showAuthors, underline=0)self.view.add_command(label="Statistics",

command=statistics, underline=0)self.run.add_command(label="Run bank11 model",

command=run, underline=0)

self.params = Parameters(endtime=2000,sourceseed=1133,counterseed=3939393,numberCustomers=50,interval=10.0,trace=0)

root = Tk()gui = MyGUI(root, title="SimPy GUI example", doc=__doc__, consoleHeight=40)gui.mainloop()



Bank Customers using SimulationStep: SimGUIStep.py

(broken for python 2.x global name ‘simulateStep’ is not defined)

Another modification of the bank11 simulation this time showing the ability to step between events.

from SimPy.SimGUI import *from SimPy.SimulationStep import *from random import Random

__version__ = '$Revision$ $Date$ kgm'

if __name__ == '__main__':


def __init__(self, seed=333):Process.__init__(self)self.SEED = seed


c = Customer(name="Customer{0:02d}".format(i,))activate(c, c.visit(timeInBank=12.0))t = rv.expovariate(1.0 / interval)yield hold, self, t



def __init__(self, name):Process.__init__(self)self.name = name

def visit(self, timeInBank=0):arrive = now()Qlength = [NoInSystem(counter[i]) for i in range(Nc)]# print("{0:7.4f} {1}: Here I am. {2} ".format(now(),# self.name,# Qlength))for i in range(Nc):

if Qlength[i] == 0 or Qlength[i] == min(Qlength):join = ibreak

yield request, self, counter[join]wait = now() - arrivewaitMonitor.observe(wait, t=now()) # Lmond# print("{0:7.4f} {1}: Waited {2:6.3f}".format(now(),# self.name,wait))tib = counterRV.expovariate(1.0 / timeInBank)yield hold, self, tib



yield release, self, counter[join]serviceMonitor.observe(now() - arrive, t=now())if trace:

gui.writeConsole("Customer leaves at {0:.1f}".format(now()))# print("{0:7.4f} {1}: Finished ".format(now(),self.name))

def showtime():gui.topconsole.config(text="time = {0}".format(now()))gui.root.update()

def runStep():if gui.noRunYet:

showwarning("SimPy warning", "Run 'Start run (stepping)' first")return

showtime()a = simulateStep(until=gui.params.endtime)if a[1] == "notResumable":

gui.writeConsole(text="Run ended. Status: {0}".format(a[0]))showtime()

def runNoStep():showtime()for i in range(gui.params.nrRuns):

simulate(until=gui.param.sendtime)showtime()gui.writeConsole("{0} simulation run(s) completed\n".format(i + 1))

def contStep():return

def model():global Nc, counter, counterRV, waitMonitor, serviceMonitor, traceglobal lastLeave, noRunYet, initializedcounterRV = Random(gui.params.counterseed)sourceseed = gui.params.sourceseednrRuns = gui.params.nrRunslastLeave = 0gui.noRunYet = Truefor runNr in range(nrRuns):

gui.noRunYet = Falsetrace = gui.params.traceif trace:

gui.writeConsole(text='\n** Run {0}'.format(runNr + 1))Nc = 2counter = [Resource(name="Clerk0"), Resource(name="Clerk1")]gui.waitMoni = waitMonitor = Monitor(name='Waiting Times')waitMonitor.xlab = 'Time'waitMonitor.ylab = 'Customer waiting time'gui.serviceMoni = serviceMonitor = Monitor(name='Service Times')serviceMonitor.xlab = 'Time'serviceMonitor.ylab = 'Total service time = wait+service'initialize()source = Source(seed=sourceseed)activate(source, source.generate(

gui.params.numberCustomers, gui.params.interval), 0.0)simulate(showtime, until=gui.params.endtime)showtime()lastLeave += now()



gui.writeConsole("{0} simulation run(s) completed\n".format(nrRuns))gui.writeConsole("Parameters:\n{0}".format(gui.params))

def modelstep():global Nc, counter, counterRV, waitMonitor, serviceMonitor, traceglobal lastLeave, noRunYetcounterRV = Random(gui.params.counterseed)sourceseed = gui.params.sourceseednrRuns = gui.params.nrRunslastLeave = 0gui.noRunYet = Truetrace = gui.params.traceif trace:

gui.writeConsole(text='\n** Run {0}'.format(runNr + 1))Nc = 2counter = [Resource(name="Clerk0"), Resource(name="Clerk1")]gui.waitMoni = waitMonitor = Monitor(name='Waiting Times')waitMonitor.xlab = 'Time'waitMonitor.ylab = 'Customer waiting time'gui.serviceMoni = serviceMonitor = Monitor(name='Service Times')serviceMonitor.xlab = 'Time'serviceMonitor.ylab = 'Total service time = wait+service'initialize()source = Source(seed=sourceseed)activate(source, source.generate(

gui.params.numberCustomers, gui.params.interval), 0.0)simulateStep(until=gui.params.endtime)gui.noRunYet = False


showwarning(title='SimPy warning',message="Run simulation first -- no data available.")

returnaver = lastLeave / gui.params.nrRunsgui.writeConsole(text="Average time for {0} customers to get through "

"bank: {1:.1f}\n({2} runs)\n".format(gui.params.numberCustomers, aver,

gui.params.nrRuns))

__doc__ = """Modified bank11.py (from Bank Tutorial) with GUI.

Model: Simulate customers arrivingat random, using a Source, requesting servicefrom two counters each with their own queuerandom servicetime.

Uses Monitor objects to record waiting timesand total service times."""




SimGUI.__init__(self, win, **p)



self.help.add_command(label="Author(s)",command=showAuthors, underline=0)

self.view.add_command(label="Statistics",command=statistics, underline=0)

self.run.add_command(label="Start run (event stepping)",command=modelstep, underline=0)

self.run.add_command(label="Next event",command=runStep, underline=0)

self.run.add_command(label="Complete run (no stepping)",command=model, underline=0)

root = Tk()gui = MyGUI(root, title="SimPy GUI example", doc=__doc__)gui.params = Parameters(endtime=2000,

sourceseed=1133,counterseed=3939393,numberCustomers=50,interval=10.0,trace=0,nrRuns=1)

gui.mainloop()

Plot

Patisserie Francaise bakery: bakery.py, bakery_OO.py

The Patisserie Francaise bakery has three ovens baking their renowned baguettes for retail and restaurant customers.They start baking one hour before the shop opens and stop at closing time.

They bake batches of 40 breads at a time, taking 25..30 minutes (uniformly distributed) per batch. Retail customers ar-rive at a rate of 40 per hour (exponentially distributed). They buy 1, 2 or 3 baguettes with equal probability. Restaurantbuyers arrive at a rate of 4 per hour (exponentially dist.). They buy 20,40 or 60 baguettes with equal probability.

The simulation answers the following questions:

a) What is the mean waiting time for retail and restaurant buyers?

2. What is their maximum waiting time?

3. What percentage of customer has to wait longer than 15 minutes?

SimPy.SimPlot is used to graph the number of baguettes over time. (KGM)

"""bakery.pyScenario:The Patisserie Francaise bakery has three ovens baking their renownedbaguettes for retail and restaurant customers. They start baking onehour before the shop opens and stop at closing time.They bake batches of 40 breads at a time,taking 25..30 minutes (uniformly distributed) per batch. Retail customersarrive at a rate of 40 per hour (exponentially distributed). They buy1, 2 or 3 baguettes with equal probability. Restaurant buyers arriveat a rate of 4 per hour (exponentially dist.). They buy 20,40 or 60baguettes with equal probability.Simulate this operation for 100 days of 8 hours shop opening time.a) What is the mean waiting time for retail and restaurant buyers?b) What is their maximum waiting time?b) What percentage of customer has to wait longer than 15 minutes??



c) Plot the number of baguettes over time for an arbitrary day.(use PLOTTING=True to do this)

"""from SimPy.Simulation import *from SimPy.SimPlot import *import random# Model components

class Bakery:def __init__(self, nrOvens, toMonitor):

self.stock = Level(name="baguette stock", monitored=toMonitor)for i in range(nrOvens):

ov = Oven()activate(ov, ov.bake(capacity=batchsize, bakery=self))

class Oven(Process):def bake(self, capacity, bakery):

while now() + tBakeMax < tEndBake:yield hold, self, r.uniform(tBakeMin, tBakeMax)yield put, self, bakery.stock, capacity

class Customer(Process):def buyBaguette(self, cusType, bakery):

tIn = now()yield get, self, bakery.stock, r.choice(buy[cusType])waits[cusType].append(now() - tIn)

class CustomerGenerator(Process):def generate(self, cusType, bakery):

while True:yield hold, self, r.expovariate(1.0 / tArrivals[cusType])if now() < (tShopOpen + tBeforeOpen):

c = Customer(cusType)activate(c, c.buyBaguette(cusType, bakery=bakery))

# Model

def model():toMonitor = Falseinitialize()if day == (nrDays - 1):

toMoni = Trueelse:

toMoni = Falseb = Bakery(nrOvens=nrOvens, toMonitor=toMoni)for cType in ["retail", "restaurant"]:

cg = CustomerGenerator()activate(cg, cg.generate(cusType=cType, bakery=b), delay=tBeforeOpen)

simulate(until=tBeforeOpen + tShopOpen)return b

# Experiment data



nrOvens = 3batchsize = 40 # nr baguettestBakeMin = 25 / 60.tBakeMax = 30 / 60. # hourstArrivals = {"retail": 1.0 / 40, "restaurant": 1.0 / 4} # hoursbuy = {"retail": [1, 2, 3], "restaurant": [20, 40, 60]} # nr baguettestShopOpen = 8tBeforeOpen = 1tEndBake = tBeforeOpen + tShopOpen # hoursnrDays = 100r = random.Random(12371)PLOTTING = True# Experimentwaits = {}waits["retail"] = []waits["restaurant"] = []for day in range(nrDays):

bakery = model()# Analysis/outputprint("bakery")for cType in ["retail", "restaurant"]:

print("Average wait for {0} customers: {1:4.2f} hours".format(cType, (1.0 * sum(waits[cType])) / len(waits[cType])))

print("Longest wait for {0} customers: {1:4.1f} hours".format(cType, max(waits[cType])))

nrLong = len([1 for x in waits[cType] if x > 0.25])nrCust = len(waits[cType])print("Percentage of {0} customers having to wait for more than"

" 0.25 hours: {1}".format(cType, 100 * nrLong / nrCust))

if PLOTTING:plt = SimPlot()plt.plotStep(bakery.stock.bufferMon,

title="Number of baguettes in stock during arbitrary day",color="blue")

plt.mainloop()


"""bakery_OO.pyScenario:The Patisserie Francaise bakery has three ovens baking their renownedbaguettes for retail and restaurant customers. They start baking onehour before the shop opens and stop at closing time.They bake batches of 40 breads at a time,taking 25..30 minutes (uniformly distributed) per batch. Retail customersarrive at a rate of 40 per hour (exponentially distributed). They buy1, 2 or 3 baguettes with equal probability. Restaurant buyers arriveat a rate of 4 per hour (exponentially dist.). They buy 20,40 or 60baguettes with equal probability.Simulate this operation for 100 days of 8 hours shop opening time.a) What is the mean waiting time for retail and restaurant buyers?b) What is their maximum waiting time?b) What percentage of customer has to wait longer than 15 minutes??c) Plot the number of baguettes over time for an arbitrary day.

(use PLOTTING=True to do this)



"""from SimPy.Simulation import *from SimPy.SimPlot import *import random# Model components ------------------------

class Bakery:def __init__(self, nrOvens, toMonitor, sim):

self.stock = Level(name="baguette stock", monitored=toMonitor, sim=sim)for i in range(nrOvens):

ov = Oven(sim=sim)sim.activate(ov, ov.bake(capacity=batchsize, bakery=self))

class Oven(Process):def bake(self, capacity, bakery):

while self.sim.now() + tBakeMax < tEndBake:yield hold, self, r.uniform(tBakeMin, tBakeMax)yield put, self, bakery.stock, capacity

class Customer(Process):def buyBaguette(self, cusType, bakery):

tIn = self.sim.now()yield get, self, bakery.stock, r.choice(buy[cusType])waits[cusType].append(self.sim.now() - tIn)

class CustomerGenerator(Process):def generate(self, cusType, bakery):

while True:yield hold, self, r.expovariate(1.0 / tArrivals[cusType])if self.sim.now() < (tShopOpen + tBeforeOpen):

c = Customer(cusType, sim=self.sim)self.sim.activate(c, c.buyBaguette(cusType, bakery=bakery))

# Model -----------------------------------

class BakeryModel(Simulation):def run(self):

# toMonitor=Falseself.initialize()toMoni = day == (nrDays - 1)b = Bakery(nrOvens=nrOvens, toMonitor=toMoni, sim=self)for cType in ["retail", "restaurant"]:

cg = CustomerGenerator(sim=self)self.activate(cg, cg.generate(

cusType=cType, bakery=b), delay=tBeforeOpen)self.simulate(until=tBeforeOpen + tShopOpen)return b

# Experiment data -------------------------nrOvens = 3batchsize = 40 # nr baguettestBakeMin = 25 / 60.tBakeMax = 30 / 60. # hours



tArrivals = {"retail": 1.0 / 40, "restaurant": 1.0 / 4} # hoursbuy = {"retail": [1, 2, 3], "restaurant": [20, 40, 60]} # nr baguettestShopOpen = 8tBeforeOpen = 1tEndBake = tBeforeOpen + tShopOpen # hoursnrDays = 100r = random.Random(12371)PLOTTING = True# Experiment ------------------------------waits = {}waits["retail"] = []waits["restaurant"] = []bakMod = BakeryModel()for day in range(nrDays):

bakery = bakMod.run()# Analysis/output -------------------------print('bakery_OO')for cType in ["retail", "restaurant"]:

print("Average wait for {0} customers: {1:4.2f} hours".format(cType, (1.0 * sum(waits[cType])) / len(waits[cType])))

print("Longest wait for {0} customers: {1:4.1f} hours".format(cType, max(waits[cType])))

nrLong = len([1 for x in waits[cType] if x > 0.25])nrCust = len(waits[cType])print("Percentage of {0} customers having to wait for more than"

" 0.25 hours: {1}".format(cType, 100 * nrLong / nrCust))

if PLOTTING:plt = SimPlot()plt.plotStep(bakery.stock.bufferMon,

title="Number of baguettes in stock during arbitrary day",color="blue")

plt.mainloop()

Bank Customers Demos SimPlot: bank11Plot.py

A modification of the bank11 simulation with graphical output. It plots service and waiting times.

""" bank11.py: Simulate customers arrivingat random, using a Source, requesting servicefrom two counters each with their own queuerandom servicetime.Uses a Monitor object to record waiting times

"""from SimPy.Simulation import *from SimPy.SimPlot import *from random import Random

class Bank11(SimPlot):def __init__(self, **p):

SimPlot.__init__(self, **p)

def NoInSystem(self, R):""" The number of customers in the resource R



in waitQ and active Q"""return (len(R.waitQ) + len(R.activeQ))

def model(self):self.counterRV = Random(self.params["counterseed"])self.sourceseed = self.params["sourceseed"]nrRuns = self.params["nrRuns"]self.lastLeave = 0self.noRunYet = Truefor runNr in range(nrRuns):

self.noRunYet = Falseself.Nc = 2self.counter = [Resource(name="Clerk0", monitored=False),

Resource(name="Clerk1", monitored=False)]self.waitMonitor = Monitor(name='Waiting Times')self.waitMonitor.xlab = 'Time'self.waitMonitor.ylab = 'Waiting time'self.serviceMonitor = Monitor(name='Service Times')self.serviceMonitor.xlab = 'Time'self.serviceMonitor.ylab = 'wait+service'initialize()source = Source(self, seed=self.sourceseed * 1000)activate(source, source.generate(self.params["numberCustomers"],

self.params["interval"]), 0.0)simulate(until=self.params['endtime'])self.lastLeave += now()

print("{0} run(s) completed".format(nrRuns))print("Parameters:\n{0}".format(self.params))


def __init__(self, modInst, seed=333):Process.__init__(self)self.modInst = modInstself.SEED = seed


c = Customer(self.modInst, name="Customer{0:02d}".format(i))activate(c, c.visit(timeInBank=12.0))t = rv.expovariate(1.0 / interval)yield hold, self, t


def __init__(self, modInst, **p):Process.__init__(self, **p)self.modInst = modInst

def visit(self, timeInBank=0):arrive = now()Qlength = [self.modInst.NoInSystem(self.modInst.counter[i])

for i in range(self.modInst.Nc)]



for i in range(self.modInst.Nc):if Qlength[i] == 0 or Qlength[i] == min(Qlength):

join = ibreak

yield request, self, self.modInst.counter[join]wait = now() - arriveself.modInst.waitMonitor.observe(wait, t=now())# print("{0:7.4f} {1}: Waited {2:6.3f}".format(now(),self.name,wait))tib = self.modInst.counterRV.expovariate(1.0 / timeInBank)yield hold, self, tibyield release, self, self.modInst.counter[join]self.modInst.serviceMonitor.observe(now() - arrive, t=now())

root = Tk()plt = Bank11()plt.params = {"endtime": 2000,

"sourceseed": 1133,"counterseed": 3939393,"numberCustomers": 50,"interval": 10.0,"trace": 0,"nrRuns": 1}

plt.model()plt.plotLine(plt.waitMonitor, color='blue', width=2)plt.plotLine(plt.serviceMonitor, color='red', width=2)root.mainloop()

Debugger

Stepping thru Simpulation Events: SimpleDebugger.py

A utility module for stepping through the events of a simulation under user control, using SimulationTrace.

"""A utility module for stepping through the events of a simulationunder user control.

REQUIRES SimPy 2.1"""from SimPy.SimulationTrace import *

from sys import version_info # Needed to determine if Python 2 or 3py_ver = version_info[0]

def stepper():evlist = Globals.sim._timestampswhile True:

if not evlist:print("No more events.")break

tEvt = evlist[0][0]who = evlist[0][2]while evlist[0][3]: # skip cancelled event notices

step()



print("\nTime now: {0}, next event at: {1} for process: {2} ".format(now(), tEvt, who.name)) # peekAll()[0],peekAll()[1].name)

if(py_ver > 2): # python3cmd = input(

"'s' next event,'r' run to end,'e' to end run, ""<time> skip to event at <time>, 'l' show eventlist, ""'p<name>' skip to event for <name>: ")

else: # python2cmd = raw_input(


try:nexttime = float(cmd)while peek() < nexttime:

step()except:

if cmd == 's':step()

elif cmd == 'r':break

elif cmd == 'e':stopSimulation()break

elif cmd == 'l':print("Events scheduled: \n{0}".format(allEventNotices()))

elif cmd[0] == 'p':while evlist and evlist[0][2].name != cmd[1:]:

step()else:

print("{0} not a valid command".format(cmd))

if __name__ == "__main__":import random as rr.seed(1234567)

class Test(Process):def run(self):

while now() < until:yield hold, self, r.uniform(1, 10)

class Waiter(Process):def run(self, evt):

def gt30():return now() > 30

yield waituntil, self, gt30print("now() is past 30")stopSimulation()

until = 100initialize()evt = SimEvent()t = Test("Test1")activate(t, t.run())t2 = Test("Test2")



activate(t2, t2.run())w = Waiter("Waiter")activate(w, w.run(evt=evt))stepper()


"""A utility module for stepping through the events of a simulationunder user control.

REQUIRES SimPy 2.1"""from SimPy.SimulationTrace import *

from sys import version_info # Needed to determine if using Python 2 or 3py_ver = version_info[0]

def stepper(whichsim):evlist = whichsim._timestampswhile True:

if not evlist:print("No more events.")break

tEvt = evlist[0][0]who = evlist[0][2]while evlist[0][3]: # skip cancelled event notices

whichsim.step()print("\nTime now: {0}, next event at: {1} for process: {2} ".format(

whichsim.now(), tEvt, who.name))

if(py_ver > 2): # Python 3cmd = input(


else: # Python 2cmd = raw_input(


try:nexttime = float(cmd)while whichsim.peek() < nexttime:

whichsim.step()except:

if cmd == 's':whichsim.step()

elif cmd == 'r':break

elif cmd == 'e':stopSimulation()break

elif cmd == 'l':print("Events scheduled: \n{0}".format(

whichsim.allEventNotices()))elif cmd[0] == 'p':



while evlist and evlist[0][2].name != cmd[1:]:whichsim.step()

else:print("{0} not a valid command".format(cmd))

if __name__ == "__main__":import random as rr.seed(1234567)

class Test(Process):def run(self):

while self.sim.now() < until:yield hold, self, r.uniform(1, 10)

class Waiter(Process):def run(self, evt):

def gt30():return self.sim.now() > 30

yield waituntil, self, gt30print("now() is past 30")self.sim.stopSimulation()

until = 100s = SimulationTrace()s.initialize()evt = SimEvent(sim=s)t = Test("Test1", sim=s)s.activate(t, t.run())t2 = Test("Test2", sim=s)s.activate(t2, t2.run())w = Waiter("Waiter", sim=s)s.activate(w, w.run(evt=evt))stepper(whichsim=s)

Authors

• Tony Vignaux <[email protected]>,


• Karen Turner (updated 2012)

• Steven Kennedy (updated for Python 3 2012)

Created 2002-December

Date 2012-April

2.9 Short Manual for SimPy Classic

SimPy Classic is a free, open-source discrete-event simulation system written in Python. It provides a number oftools for programmers writing simulation programs. This document is a description of basic techniques of SimPy.It describes a subset of SimPy’s capabilities - sufficient, we think, to develop standard simulations. You may alsofind The Bank tutorial included in SimPy’s distribution helpful in the early stages. The full SimPy Classic Manual,included in the distribution, is more detailed.





You need to write Python code to develop a SimPy Classic model. In particular, you will have to define and use classesand their objects. Python is free and open-source and is available on most platforms. You can find out more aboutit and download it from the Python web-site where there is full documentation and tutorials. SimPy requires Pythonversion 2.3 or later.

Contents:

2.9.1 Introduction to SimPy Classic

Authors G A Vignaux and Klaus Muller

Date Feb 24, 2018

Release 2.3.3


Contents

• Introduction to SimPy Classic

– Introduction

– Simulation with SimPy

– Processes

– Resources

– Random Number Generation

– Monitors and Recording Simulation Results

– Appendices

Introduction

The active elements (or entities) of a SimPy model are objects of a class defined by the programmer (see Processes,section 3). Each entity has a standard method, a Process Execution Method (referred to by SimPy programmers as aPEM) which specifies its actions in detail. Each PEM runs in parallel with (and may interact with) the PEMs of otherentities.

The activity of an entity may be delayed for fixed or random times, queued at resource facilities, and may be interruptedby or interact in different ways with other entities and components. For example in a gas station model, automobileentities (objects of an Automobile Class) may have to wait at the gas station for a pump to become available. Onobtaining a pump it takes time to fill the tank. The pump is then released for the next automobile in the queue if thereis one.

SimPy has three kinds of resource facilities (Resources, Levels, and Stores). Each type models a congestion pointwhere entities queue while waiting to acquire or, in some cases, to deposit a resource.SimPy automatically handles thequeueing.

• Resources have one or more identical resource units, each of which can be held by entities. Extending theexample above, the gas station might be modelled as a Resource with its pumps as resource units. When acar requests a pump the gas station resource automatically queues it until a pump becomes available (perhapsimmediately). The car holds the pump until it finishes refuelling and then releases it for use by the next car.

2.9. Short Manual for SimPy Classic 221



• Levels (not treated here) model the supply and consumption of a homogeneous undifferentiated “material”.The Level holds an amount that is fully described by a non-negative number which can be increased or decreasedby entities. For example, a gas station stores gas in large storage tanks. The tanks can be filled by tankers andemptied by cars refuelling. In contrast to the operation of a Resource, a car need not return the gas to the gasstation.

• Stores (not treated here) model the production and consumption of distinguishable items. A Store holds a listof items. Entities can insert or remove items from the list and these can be of any type. They can even be SimPyprocess objects. For example, the gas station holds spares of different types. A car might request a set of sparesfrom the Store. The store is replenished by deliveries from a warehouse.

SimPy also supplies Monitors and Tallys to record simulation events. Monitors are used to compile summary statisticssuch as waiting times and queue lengths. These statistics includes simple averages and variances, time-weightedaverages, or histograms. In particular, data can be gathered on the queues associated with Resources, Levels andStores. For example we may collect data on the average number of cars waiting at the gas station and the distributionof their waiting times. Monitors preserve complete time-series records that may later be used for more advanced post-simulation analyses. A simpler version of a Monitor is a Tally which provides most of the statistical facilities but doesnot retain a complete time-series record. See the SimPy Manual for details on the Tally.

Simulation with SimPy

To use the SimPy simulation system in your Python program you must import its Simulation module using:


We recommend that new users instead import SimPy.SimulationTrace, which works the same but also auto-matically produces a timed listing of events as the model executes. (An example of such a trace is shown in TheResource Example with Tracing):

from SimPy.SimulationTrace import *

Discrete-event simulation programs automatically maintain the current simulation time in a software clock. Thiscannot be directly changed by the user. In SimPy the current clock value is returned by the now() function andyou can access this or print it out. At the start of the simulation it is set to 0.0. While the simulation program runs,simulation time steps forward from one event to the next. An event occurs whenever the state of the simulated systemchanges such as when a car arrives or departs from the gas station.

The initialize statement initialises global simulation variables and sets the software clock to 0.0. It must appearin your program before any SimPy process objects are activated.

initialize()

This is followed by SimPy statements creating and activating entities (that is, SimPy process objects). Activationof entities adds events to the simulation event schedule. Execution of the simulation itself starts with the followingstatement:

simulate(until=endtime)

The simulation then starts, and SimPy seeks and executes the first event in the schedule. Having executed that event,the simulation seeks and executes the next event, and so on.

Typically a simulation terminates when there are no more events to execute or when the endtime is reached but it canbe stopped at any time by the command:

stopSimulation( )

After the simulation stops, further statements can be executed. now() will retain the time of stopping and data heldin Monitors will be available for display or further analysis.

The following fragment shows only the main block in a simulation program to illustrate the general structure. Acomplete Example Program is shown later. Here Car is a Process class with a go as its PEM (described later) and c



is made an entity of that class, that is, a particular car. Activating c has the effect of scheduling at least one event bystarting c’s PEM. The simulate(until=1000.0) statement starts the simulation itself. This immediately jumpsto the first scheduled event. It will continue executing events until it runs out of events to execute or the simulationtime reaches 1000.0. When the simulation stops the Report function is called to display the results:

1 class Car(Process):2 def go(self):3 # PEM for a Car4 ...5

6 def Report():7 # print results when finished8 ...9

10 initialize()11 c = Car(name="Car23")12 activate(c, c.go(), at=0.0)13 simulate(until=1000.0)14

15 Report()

In addition to SimPy.Simulation there are three alternative simulation libraries with the same simulation capa-bilities but with special facilities. Beside SimPy.SimulationTrace, already mentioned, there are SimPy.SimulationRT for real time synchronisation and SimPy.SimulationStep for event-stepping through a simu-lation. See the SimPy Manual for more information.

Processes

SimPy’s active objects (entities) are process objects – instances of a class written by the user that inherits from SimPy’sProcess class.

For example, if we are simulating a gas station we might model each car as an object of the class Car. A car arrivesat the gas station (modelled as a Resource with pumps). It requests a pump and may need to wait for it. Then it fillsits tank and when it has finished it releases the pump. It might also buy an item from the station store. The Car classspecifies the logic of these actions in its Process Execution Method (PEM). The simulation creates a number of carsas it runs and their evolutions are directed by their Car class’s PEM.

Defining a process

Each Process class inherits from SimPy’s Process class. For example the header of the definition of a Car Processclass would be:

class Car(Process):

At least one Process Execution Method (PEM) must be defined in each Process class (though an entity can have onlyone PEM active). A PEM may have arguments in addition to the required self argument needed by all Python classmethods. Naturally, other methods and, in particular, an __init__, may be defined.

• A Process Execution Method (PEM) defines the actions that are performed by its process objects.Each PEM must contain at least one of the special ‘‘yield‘‘ statements, described later. This makes the PEM aPython generator function so that it has resumable execution – it can be restarted again after the yield statementwithout losing its current state. A PEM may have any name of your choice. For example it may be calledexecute( ) or run( ). However, if a PEM is called ACTIONS, SimPy recognises this as a PEM. This cansimplify the start method as explained below.



The yield statements are simulation commands which affect an ongoing life cycle of Process objects. Thesestatements control the execution and synchronisation of multiple processes. They can delay a process, put itto sleep, request a shared resource or provide a resource. They can add new events to the simulation eventschedule, cancel existing ones, or cause processes to wait for a change in the simulated system’s state.

For example, here is a Process Execution Method, go(self), for the simple Car class that does no more thandelay for a time. As soon as it is activated it prints out the current time, the car object’s name and the wordStarting. After a simulated delay of 100.0 time units (in the yield hold, ... statement) it announcesthat this car has “Arrived”:

def go(self):print("%s %s %s" % (now(), self.name, 'Starting'))yield hold, self, 100.0print("%s %s %s" % (now(), self.name, 'Arrived'))

A process object’s PEM starts execution when the object is activated, provided thesimulate(until=endtime) statement has been executed.

• __init__(self, . . . ), where . . . indicates other arguments. This method is optional but is useful to initialise theprocess object, setting values for its attributes. As for any sub-class in Python, the first line of this method mustcall the Process class’s __init__( ) method in the form:


You can then use additional commands to initialise attributes of the Process class’s objects. You can also overridethe standard name attribute of the object.

If present, the __init__( ) method is always called whenever you create a new process object. If you donot wish to provide for any attributes other than a name, the __init__ method may be dispensed with. Anexample of an __init__( ) method is shown in the Example Program.

Creating a process object

An entity (process object) is created in the usual Python manner by calling the Class. Process classes have a singleattribute, name which can be specified even if no __init__ method is defined. name defaults to 'a_process'unless the user specified a different one.

For example to create a new Car object with a name Car23:

c = Car(name="Car23")

Starting SimPy Process Objects

An entity (process object) is “passive” when first created, i.e., it has no events scheduled for it. It must be activated tostart its Process Execution Method. To do this you can use either the activate function or the start method ofthe Process.

activate

Activating an entity by using the SimPy activate function:

• activate(p, p.pemname([args])[,{at=t|delay=period}])

activates process object p, provides its Process Execution Method p.pemname( ) with the arguments argsand possibly assigns values to the other optional parameters. You must choose one (or neither) of at=t anddelay=period. The default is to activate at the current time (at=now( )) and with no delay (delay=0).



For example: to activate an entity, cust at time 10.0 using its PEM called lifetime:

cust = Customer()activate(cust, cust.lifetime(), at=10.0)

start

An alternative to the activate() function is the start method of Process objects:

• p.start(p.pemname([args])[,{at=t|delay=period}])

Here p is a Process object. Its PEM, pemname, can have any identifier (such as run, lifecycle, etc) and anyarguments args.

For example, to activate the process object cust using the PEM with identifier lifetime at time 10.0 wewould use:

cust.start(cust.lifetime(), at=10.0)

The standard PEM name, ACTIONS

The identifier ACTIONS is recognised by SimPy as a PEM name and can be used (or implied) in the start method.

• p.start([p.ACTIONS()] [,{at=t|delay=period}])

ACTIONS cannot have parameters. The call p.ACTIONS() is optional but may make your code clearer.

For example, to activate the Process object cust with a PEM called ACTIONS at time 10.0, the following areequivalent (and the second version more convenient):

cust.start(cust.ACTIONS(), at=10.0)cust.start(at=10.0)

A reminder: Even activated process objects will not actually start operating until the simulate() statement isexecuted.

Elapsing time in a Process

A PEM uses the yield hold command to temporarily delay a process object’s operations. This might represent aservice time for the entity. (Waiting is handled automatically by the resource facilities and is not modelled by yieldhold)

yield hold

yield hold, self, t

Causes the entity to delay t time units. After the delay, it continues with the next statement in its PEM.During the hold the entity’s operations are suspended.

Paradoxically, in the model world, the entity is considered to be busy during this simulated time. For example, it mightbe involved in filling up with gas or driving. In this state it can be interrupted by other entities.



More about Processes

An entity (Process object) can be “put to sleep” or passivated using

yield passivate, self

(and it can be reactivated by another entity using reactivate), or permanently removed from the future event queueby the command self.cancel(). Active entities can be interrupted by other entities. Examine the full SimPyManual for details.

A SimPy Program

This is a complete SimPy script. We define a Car class with a PEM called go( ). We also (for interest) define an__init__( ) method to provide individual cars with an identification name and engine size, cc. The cc attributeis not used in this very simple example.

Two cars, p1 and p2 are created. p1 and p2 are activated to start at simulation times 0.6 and 0.0, respectively. Notethat these will not start in the order they appear in the program listing. p2 actually starts first in the simulation. Nothinghappens until the simulate(until=200) statement at which point the event scheduler starts operating by findingthe first event to execute. When both cars have finished (at time 6.0+100.0=106.0) there will be no more eventsso the simulation will stop:

from SimPy.Simulation import Process, activate, hold, initialize, now, simulate





Running this program gives the following output:

0 Car2 Starting6.0 Car1 Starting100.0 Car2 Arrived106.0 Car1 ArrivedCurrent time is 106.0

If, instead one chose to import SimPy.SimulateTrace at the start of the program one would obtain the followingoutput. (The meaning of the phrase prior : False in the first two lines is described in the full SimPy Manual.



prior is an advanced technique for fine control of PEM priorities but seldom affects simulated operations and sonormally can be ignored/)

0 activate <Car1> at time: 6.0 prior: False0 activate <Car2> at time: 0 prior: False0 Car2 Starting0 hold <Car2> delay: 100.06.0 Car1 Starting6.0 hold <Car1> delay: 100.0100.0 Car2 Arrived100.0 <Car2> terminated106.0 Car1 Arrived106.0 <Car1> terminatedCurrent time is 106.0

Resources

The three resource facilities provided by SimPy are Resources, Levels and Stores. Each models a congestionpoint where process objects may have to queue up to access resources. This section describes the Resource type ofresource facility. Levels and Stores are not treated in this introduction.

An example of queueing for a Resource might be a manufacturing plant in which a Task (modelled as an entityor Process object) needs work done by a Machine (modelled as a Resource object). If all of the Machines arecurrently being used, the Task must wait until one becomes free. A SimPy Resource can have a number of identicalunits, such as a number of identical machine units. An entity obtains a unit of the Resource by requesting itand, when it is finished, releasing it. A Resource maintains a list (the waitQ) of entities that have requested butnot yet received one of the Resource’s units, and another list (the activeQ) of entities that are currently using a unit.SimPy creates and updates these queues itself – the user can read their values, but should not change them.

Defining a Resource object

A Resource object, r, is established by the following statement:

r = Resource(capacity=1,name='a_resource',unitName='units',monitored=False)

where

• capacity (positive integer) specifies the total number of identical units in Resource object r.

• name (string) the name for this Resource object (e.g., 'gasStation').

• unitName (string) the name for a unit of the resource (e.g., 'pump').

• monitored (False or True) If set to True, then information is gathered on the sizes of r’s waitQ andactiveQ, otherwise not.

For example, in the model of a 2-pump gas-station we might define:

gasstation = Resource(capacity=2, name='gasStation', unitName='pump')

Each Resource object, r, has the following additional attributes:

• r.n, the number of units that are currently free.



• r.waitQ, a queue (list) of processes that have requested but not yet received a unit of r, so len(r.waitQ)is the number of process objects currently waiting.

• r.activeQ, a queue (list) of process objects currently using one of the Resource’s units, so len(r.activeQ) is the number of units that are currently in use.

• r.waitMon, the record (made by a Monitor whenever monitored == True) of the activity in r.waitQ. So, for example, r.waitMon.timeaverage() is the average number of processes in r.waitQ.See Data Summaries for an example.

• r.actMon, the record (made by a Monitor whenever monitored==True) of the activity in r.activeQ.

Requesting and releasing a unit of a Resource

A process can request and later release a unit of the Resource object, r, by using the following yield commands in aProcess Execution Method:

yield request

• yield request,self,r

Requests a unit of Resource r

If a Resource unit is free when the request is made, the requesting entity takes it and moves on to the next statementin its PEM. It is added to the Resource’s activeQ.

If no Resource unit is available when the request is made, the requesting entity is appended to the Resource’s waitQand suspended. The next time a unit becomes available the first entity in the r.waitQ takes it and continues itsexecution.

For example, a Car might request a pump:

yield request, self, gasstation

(It is actually requesting a unit of the gasstation, i.e. a pump.) An entity holds a resource unit until it releases it.

Entities can use a priority system for queueing. They can also preempt (that is, interrupt) others already in the system.They can also renege from the waitQ (that is, abandon the queue if it takes too long). This is achieved by an extensionto the yield request command. See the main SimPy Manual.

yield release

yield release, self, r

Releases the unit of r.

The entity is removed from r.activeQ and continues with its next statement. If, when the unit of r is released,another entity is waiting (in waitQ) it will take the unit, leave the waitQ and move into the activeQ and go onewith its PEM.

For example the Car might release the pump unit of the gasstation:

yield release, self, gasstation



Resource Example

In this complete script, the gasstation Resource object is given two resource units (capacity=2). Four carsarrive at the times specified in the program (not in the order they are listed). They all request a pump and use it for100 time units:

1 from SimPy.Simulation import (Process, Resource, activate, initialize, hold,2 now, release, request, simulate)3

4

5 class Car(Process):6 def __init__(self, name, cc):7 Process.__init__(self, name=name)8 self.cc = cc9

10 def go(self):11 print('%s %s %s' % (now(), self.name, 'Starting'))12 yield request, self, gasstation13 print('%s %s %s' % (now(), self.name, 'Got a pump'))14 yield hold, self, 100.015 yield release, self, gasstation16 print('%s %s %s' % (now(), self.name, 'Leaving'))17

18

19 gasstation = Resource(capacity=2, name='gasStation', unitName='pump')20 initialize()21 c1 = Car('Car1', 2000)22 c2 = Car('Car2', 1600)23 c3 = Car('Car3', 3000)24 c4 = Car('Car4', 1600)25 activate(c1, c1.go(), at=4.0) # activate at time 4.026 activate(c2, c2.go()) # activate at time 0.027 activate(c3, c3.go(), at=3.0) # activate at time 3.028 activate(c4, c4.go(), at=3.0) # activate at time 2.029 simulate(until=300)30 print('Current time is %s' % now())

This program results in the following output:

0 Car2 Starting0 Car2 Got a pump3.0 Car3 Starting3.0 Car3 Got a pump3.0 Car4 Starting4.0 Car1 Starting100.0 Car2 Leaving100.0 Car4 Got a pump103.0 Car3 Leaving103.0 Car1 Got a pump200.0 Car4 Leaving203.0 Car1 LeavingCurrent time is 203.0

And, if we use SimPy.SimulationTrace to get an automatic trace we get the result shown in Appendix TheResource Example with Tracing. (It is rather long to be inserted here).



Random Number Generation

Simulations usually need random numbers. By design, SimPy does not provide its own random number generators, sousers need to import them from some other source. Perhaps the most convenient is the standard Python random module.It can generate random variates from the following continuous distributions: uniform, beta, exponential, gamma,normal, log-normal, Weibull, and vonMises. It can also generate random variates from some discrete distributions.Consult the module’s documentation for details. Excellent brief descriptions of these distributions, and many others,can be found in the Wikipedia.

Python’s random module can be used in two ways: you can import the methods directly or you can import theRandom class and make your own random objects. In the second method, each object gives a different randomnumber sequence, thus providing multiple random streams as in some other simulation languages such as Simscriptand ModSim.

Here the first method is illustrated. A single pseudo-random sequence is used for all calls. You import the methodsyou need from the random module. For example:

from random import seed, random, expovariate, normalvariate

In simulation it is good practice to set the initial seed for the pseudo-random sequence at the start of each run. Youthen have good control over the random numbers used. Replications and comparisons are easier and, together withvariance reduction techniques, can provide more accurate estimates. In the following code snippet we set the initialseed to 333555. X and Y are pseudo-random variates from the two distributions. Both distributions have the samemean:

1 from random import seed, expovariate, normalvariate2

3 seed(333555)4 X = expovariate(0.1)5 Y = normalvariate(10.0, 1.0)

Monitors and Recording Simulation Results

A Monitor enables us to observe a particular variable of interest and to hold a time series of its values. It can return asimple data summary either during or at the completion of a simulation run.

For example we might use one Monitor to record the waiting time for each of a series of customers and another torecord the total number of customers in the shop. In a discrete-event system the number of customers changes only atarrival or departure events and it is at those events that the number in the shop must be observed. A Monitor providessimple statistics useful either alone or as the start of a more sophisticated statistical analysis.

When Resources are defined, a Monitor can be set up to automatically observe the lengths of each of their queues.

The simpler version, the Tally, is not covered in this document. See the SimPy Manual for more details.

Defining Monitors

The Monitor class preserves a complete time-series of the observed data values, y, and their associated times, t.The data are held in a list of two-item sub-lists, [t,y]. Monitors calculate data summaries using this time-serieswhen your program requests them. In long simulations their memory demands may be a disadvantage

To define a new Monitor object:

• m = Monitor(name='a_Monitor')

where name is a descriptive name for the Monitor object. The descriptive name is used when the data is graphedor tabulated.



https://en.wikipedia.org/wiki/Main_Page


For example, to record the waiting times of cars in the gas station we might use a Monitor:

waittimes = Monitor(name='Waiting times')

Observing data

Monitors use their observe method to record data. Here and in the next section, m is a Monitor object:

• m.observe(y [,t])

records the current value of the variable, y and time t (or the current time, now(), if t is missing). A Monitorretains the two values as a sub-list [t,y].

For example, using the Monitor in the previous example, we might record the waiting times of the cars as shown inthe following fragment of the PEM of a Car:

1 startwaiting = now() # start wait2 yield request, self, gasstation3 waitingtime = now() - startwaiting # time spent waiting4

5 waittimes.observe(waitingtime)

The first three lines measure the waiting time (from the time of the request to the time the pump is obtained). The lastrecords the waiting time in the waittimes Monitor.

The data recording can be reset to start at any time in the simulation:

• m.reset([t])

resets the observations. The recorded data is re-initialised, and the observation starting time is set to t, or to thecurrent simulation time, now( ), if t is missing.

Data summaries

The following simple data summaries are available from Monitors at any time during or after the simulation run:

• m[i] holds the i-th observation as a two-item list, [ti, yi]

• m.yseries( ) is a list of the recorded data values, yi

• m.tseries( ) is a list of the recorded times, ti

• m.count( ), the current number of observations. (This is the same as len(r)).

• m.total( ), the sum of the y values

• m.mean( ), the simple numerical average of the observed y values, ignoring the times at which they weremade. This is m.total( )/m.count( ).

• m.var( ) the sample variance of the observations, ignoring the times at which they were made. If an unbiasedestimate of the population variance is desired, the sample variance should be multiplied by n/(n-1), where n =m.count( ). In either case the standard deviation is, of course, the square-root of the variance

• m.timeAverage([t]) the time-weighted average of y, calculated from time 0 (or the last time m.reset([t]) was called) to time t (or to the current simulation time, now(), if t is missing).

This is intended to measure the average of a quantity that always exists, such as the length of a queue orthe amount in a Level1. In discrete-event simulation such quantity changes are always instantaneous jumps

1 timeAverage is not intended to measure instantaneous values such as a service time or a waiting time. m.mean() is used for that.



��

�

��

��

��

t3t2t1

m.mean()

t4 t5

y3y2

y1

y4

y5

n=5

0

Fig. 2.4: m.mean is the simple average of the y values observed.

occurring at events. The recorded times and new levels are sufficient information to calculate the average overtime. The graph shown in the figure below illustrates the calculation. The total area under the line is calculatedand divided by the total time of observation. For accurate time-average results y must be piecewise constant likethis and observed just after each change in its value. That is, the y value observed must be the new value afterthe state change.

��

��

��

��

��

t3t2t1 t4 t5

y1y2

y3y4

y5m.timeaverage()

0 t

Fig. 2.5: m.timeAverage([t]) is the time-weighted average of the observed y values. Each y value is weightedby the time for which it exists. The average is the area under the above curve divided by the total time, t.

• m.timeVariance([t]) the time-weighted variance of the y values calculated from time 0 (or the last timem.reset([t]) was called) to time t (or to the current simulation time, now(), if t is missing).

• m.__str__( ) is a string that briefly describes the current state of the monitor. This can be used in a printstatement.

Monitoring Resource Queues

If a Resource, m, (and similarly for a Level or a Store) is defined with monitored=True, SimPy automaticallyrecords the lengths of its associated queues (i.e. waitQ and activeQ for Resources, and the analogous queues forLevels and Stores). These records are kept in Monitors m.waitMon for the waitQ and m.actMon for the activeQ(and analogously for the other resource types). This solves a problem, particularly for the waitQ which cannot easilybe recorded externally to the resource.

Complete time series for queue lengths are maintained and can be used for advanced post-simulation statistical analy-ses and to display summary statistics.

More on Monitors

When a Monitor is defined it is automatically entered into a global list allMonitors. Each Monitor also hasa descriptive label for its variable values, y, and their corresponding times, t, that can be used when the data areplotted. The function startCollection() can be called to initialise all the Monitors in allMonitors ata certain simulation time. This is helpful when a simulation needs a ‘warmup’ period to achieve steady state before



measurements are started. A Monitor can also generate a Histogram of the data (The SimPy Tally is an alternativeto Monitor that does essentially the same job but uses less storage space at some cost of speed and flexibility. Seethe SimPy Manual for more information on these options.)

Appendices

The Resource Example with Tracing

This is the trace produced in the Resource example when SimPy.SimulationTrace is imported at the head ofthe program. The relevance of the phrases `prior: False` and `priority: default` refer to advancedbut seldom-needed methods for fine-grained control of event timing, as explained in the SimPy Manual. The tracecontains the results of all the print output specifically called for by the user’s program (for example, line 5) but addsa line for every event executed. (The request at time 3.0, for example, lists the contents of the waitQ and theactiveQ for the gasstation.)

1 0 activate <Car1> at time: 4.0 prior: False2 0 activate <Car2> at time: 0 prior: False3 0 activate <Car3> at time: 3.0 prior: False4 0 activate <Car4> at time: 3.0 prior: False5 0 Car2 Starting6 0 request <Car2> <gasStation> priority: default7 . . .waitQ: []8 . . .activeQ: ['Car2']9 0 Car2 Got a pump

10 0 hold <Car2> delay: 100.011 3.0 Car3 Starting12 3.0 request <Car3> <gasStation> priority: default13 . . .waitQ: []14 . . .activeQ: ['Car2', 'Car3']15 3.0 Car3 Got a pump16 3.0 hold <Car3> delay: 100.017 3.0 Car4 Starting18 3.0 request <Car4> <gasStation> priority: default19 . . .waitQ: ['Car4']20 . . .activeQ: ['Car2', 'Car3']21 4.0 Car1 Starting22 4.0 request <Car1> <gasStation> priority: default23 . . .waitQ: ['Car4', 'Car1']24 . . .activeQ: ['Car2', 'Car3']25 100.0 reactivate <Car4> time: 100.0 prior: 126 100.0 release <Car2> <gasStation>27 . . .waitQ: ['Car1']28 . . .activeQ: ['Car3', 'Car4']29 100.0 Car2 Leaving30 100.0 <Car2> terminated31 100.0 Car4 Got a pump32 100.0 hold <Car4> delay: 100.033 103.0 reactivate <Car1> time: 103.0 prior: 134 103.0 release <Car3> <gasStation>35 . . .waitQ: []36 . . .activeQ: ['Car4', 'Car1']37 103.0 Car3 Leaving38 103.0 <Car3> terminated39 103.0 Car1 Got a pump40 103.0 hold <Car1> delay: 100.041 200.0 release <Car4> <gasStation>



42 . . .waitQ: []43 . . .activeQ: ['Car1']44 200.0 Car4 Leaving45 200.0 <Car4> terminated46 203.0 release <Car1> <gasStation>47 . . .waitQ: []48 . . .activeQ: []49 203.0 Car1 Leaving50 203.0 <Car1> terminated51 Current time is 203.0

2.9.2 Cheatsheets

If you want to have a list of all SimPy Classic commands, their syntax and parameters for your desktop, print out acopy of the SimPy Classic Cheatsheet, or better yet bookmark them in your favorite browser.

The Cheatsheet comes as a PDF or Excel:

• PDF A4 page

• Excel A4 page

2.9.3 Acknowledgments

SimPy 2.3.3 has been primarily developed by Stefan Scherfke and Ontje Lünsdorf, starting from SimPy 1.9. Theirwork has resulted in a most elegant combination of an object oriented API with the existing API, maintaining full back-ward compatibility. It has been quite easy to integrate their product into the existing SimPy code and documentationenvironment.

Thanks, guys, for this great job! SimPy 2.0 is dedicated to you!

The many contributions of the SimPy user and developer communities are of course also gratefully acknowledged.

2.9.4 Indices and tables

• genindex

• search


../../_static/cheatsheet_2_3.xls

CHAPTER 3

SimPy Classic Tutorials

There are two styles of modelling using SimPy. The first, which we call the Classical style, uses user-defined Processobjects each representing an active entity of the simulation. The simulation is started by creating one or more entities,activating their Process Execution Methods and then, in a main block of the program, calling the simulate()function to start the simulation. This simpler style seems to be more acceptable to many users.

In the second style, referred to as the OO Style, a user-defined object of a Simulation class is created and executed.This object executes the whole simulation. This is particularly suitable for running replications of the simulation andthe execution of a simulation object is independent of others.

3.1 The Bank

3.1.1 Introduction

In this tutorial, SimPy is used to develop a simple simulation of a bank with a number of tellers. This Python packageprovides Processes to model active components such as messages, customers, trucks, and planes. It has three classesto model facilities where congestion might occur: Resources for ordinary queues, Levels for the supply of quantitiesof material, and Stores for collections of individual items. Only examples of Resources are described here. It alsoprovides Monitors and Tallys to record data like queue lengths and delay times and to calculate simple averages. Ituses the standard Python random package to generate random numbers.

Starting with SimPy 2.0 an object-oriented programmer’s interface was added to the package. It is compatible withthe current procedural approach which is used in most of the models described here.

SimPy can be obtained from: https://github.com/SimPyClassic/SimPyClassic. The examples run with SimPy version1.5 and later. This tutorial is best read with the SimPy Manual or Cheatsheet at your side for reference.

Before attempting to use SimPy you should be familiar with the Python language. In particular you should be able touse classes. Python is free and available for most machine types. You can find out more about it at the Python website. SimPy is compatible with Python version 2.7 and Python 3.x.

235







3.1.2 A customer arriving at a fixed time

In this tutorial we model a simple bank with customers arriving at random. We develop the model step-by-step,starting out simply, and producing a running program at each stage. The programs we develop are available withoutline numbers and ready to go, in the bankprograms directory. Please copy them, run them and improve them - andin the tradition of open-source software suggest your modifications to the SimPy users list. Object-Oriented versionsof all the models are included in the bankprograms_OO sub directory.

A simulation should always be developed to answer a specific question; in these models we investigate how changingthe number of bank servers or tellers might affect the waiting time for customers.

We first model a single customer who arrives at the bank for a visit, looks around at the decor for a time and thenleaves. There is no queueing. First we will assume his arrival time and the time he spends in the bank are fixed.

We define a Customer class derived from the SimPy Process class. We create a Customer object, c who arrivesat the bank at simulation time 5.0 and leaves after a fixed time of 10.0 minutes.

Examine the following listing which is a complete runnable Python script. We use comments to divide the scriptup into sections. This makes for clarity later when the programs get more complicated. At #1 is a normal Pythondocumentation string; #2 imports the SimPy simulation code.

The Customer class definition at #3 defines our customer class and has the required generator method (called visit#4) having a yield statement #6. Such a method is called a Process Execution Method (PEM) in SimPy.

The customer’s visit PEM at #4, models his activities. When he arrives (it will turn out to be a ‘he’ in this model),he will print out the simulation time, now(), and his name at #5. The function now() can be used at any time in thesimulation to find the current simulation time though it cannot be changed by the programmer. The customer’s namewill be set when the customer is created later in the script at #10.

He then stays in the bank for a fixed simulation time timeInBank at #6. This is achieved by the yield hold,self,timeInBank statement. This is the first of the special simulation commands that SimPy offers.

After a simulation time of timeInBank, the program’s execution returns to the line after the yield statement at#6. The customer then prints out the current simulation time and his name at #7. This completes the declaration of theCustomer class.

The call initialize() at #9 sets up the simulation system ready to receive activate calls. At #10, we createa customer, c, with name Klaus. All SimPy Processes have a name attribute. We activate Klaus at #11specifying the object (c) to be activated, the call of the action routine (c.visit(timeInBank = 10.0)) andthat it is to be activated at time 5 (at = 5.0). This will activate Klaus exactly 5 minutes after the current time, inthis case after the start of the simulation at 0.0. The call of an action routine such as c.visit can specify the valuesof arguments, here the timeInBank.

Finally the call of simulate(until=maxTime) at #12 will start the simulation. This will run until the simulationtime is maxTime unless stopped beforehand either by the stopSimulation() command or by running out ofevents to execute (as will happen here). maxTime was set to 100.0 at #8.

""" bank01: The single non-random Customer """ # 1from SimPy.Simulation import * # 2

# Model components -----------------------------

class Customer(Process): # 3""" Customer arrives, looks around and leaves """

def visit(self, timeInBank): # 4print("%2.1f %s Here I am" % (now(), self.name)) # 5yield hold, self, timeInBank # 6print("%2.1f %s I must leave" % (now(), self.name)) # 7

236 Chapter 3. SimPy Classic Tutorials


# Experiment data ------------------------------

maxTime = 100.0 # minutes #8timeInBank = 10.0 # minutes


initialize() # 9c = Customer(name="Klaus") # 10activate(c, c.visit(timeInBank), at=5.0) # 11simulate(until=maxTime) # 12

The short trace printed out by the print statements shows the result. The program finishes at simulation time 15.0because there are no further events to be executed. At the end of the visit routine, the customer has no more actionsand no other objects or customers are active.

5.0 Klaus Here I am15.0 Klaus I must leave

3.1.3 The Bank in object-oriented style

Now look at the same model developed in object-oriented style. As before Klaus arrives at the bank for a visit, looksaround at the decor for a time and then leaves. There is no queueing. The arrival time and the time he spends in thebank are fixed.

The key point is that we create a Simulation object and run that. In the program at #1 we import a new class,Simulation together with the familiar Process class, and the hold verb. Here we are using the recommendedexplicit form of import rather than the deprecated import *.

Just as before, we define a Customer class, derived from the SimPy Process class, which has the required gener-ator method (PEM), here called visit.

The customer’s visit PEM models his activities. When he arrives at the bank Klaus will print out both the currentsimulation time, self.sim.now(), and his name, self.name. The prefix self.sim is a reference to thesimulation object where this customer exists, thus self.sim.now() refers to the clock for that simulation object.Every Process instance is linked to the simulation in which it is created by assigning to its sim parameter when itis created (at #4).

Now comes the major difference from the Classical SimPy program structure. We define a class BankModel fromthe Simulation class at #3. Now any instance of BankModel is an independent simulation with its own event listand its own time axis. A BankModel instance can activate processes and start the execution of a simulation on itstime axis.

In the BankModel class, we define a run method which, when executes the BankModel instance, i.e. performs asimulation experiment. When it starts it initializes the simulation with it event list and sets the time to 0.

#4 creates a Customer object and the parameter assignment sim = self ties the customer instance to this and onlythis simulation. The customer does not exist outside this simulation. The call of simulate(until=maxTime)at #5 starts the simulation. It will run until the simulation time is maxTime unless stopped beforehand either by thestopSimulation() command or by running out of events to execute (as will happen here). maxTime is set to100.0 at #6.

Note: If model classes like the BankModel are to be given any other attributes, they must have an __init__

3.1. The Bank 237


method in which these attributes are assigned. Such an __init__ method must first call Simulation.__init__(self) to also initialize the Simulation class from which the model inherits.

A new, independent simulation object, mymodel, is created at #7. Its run method is executed at #8.

""" bank01_OO: The single non-random Customer """from SimPy.Simulation import Simulation, Process, hold # 1# Model components -----------------------------


def visit(self, timeInBank):print("%2.1f %s Here I am" %

(self.sim.now(), self.name))yield hold, self, timeInBankprint("%2.1f %s I must leave" %

(self.sim.now(), self.name))

# Model ----------------------------------------

class BankModel(Simulation): # 3def run(self):

""" An independent simulation object """c = Customer(name="Klaus", sim=self) # 4self.activate(c, c.visit(timeInBank), at=tArrival)self.simulate(until=maxTime) # 5

# Experiment data ------------------------------maxTime = 100.0 # minutes #6timeInBank = 10.0 # minutestArrival = 5.0 # minutes

# Experiment -----------------------------------mymodel = BankModel() # 7mymodel.run() # 8



All of The Bank programs have been written in both the procedural and object orientated styles.

3.1.4 A customer arriving at random

Now we extend the model to allow our customer to arrive at a random simulated time though we will keep the time inthe bank at 10.0, as before.

The change occurs at #1, #2, #3, and #4 in the program. At #1 we import from the standard Python random moduleto give us expovariate to generate the random time of arrival. We also import the seed function to initialize



the random number stream to allow control of the random numbers. At #2 we provide an initial seed of 99999. Anexponential random variate, t, is generated at #3. Note that the Python Random module’s expovariate functionuses the average rate (that is, 1.0/mean) as the argument. The generated random variate, t, is used at #4 as the atargument to the activate call.

""" bank05: The single Random Customer """from SimPy.Simulation import *from random import expovariate, seed # 1


class Customer(Process):""" Customer arrives at a random time,

looks around and then leaves """

def visit(self, timeInBank):print("%f %s Here I am" % (now(), self.name))yield hold, self, timeInBankprint("%f %s I must leave" % (now(), self.name))

# Experiment data -------------------------

maxTime = 100.0 # minutestimeInBank = 10.0# Model/Experiment ------------------------------

seed(99999) # 2initialize()c = Customer(name="Klaus")t = expovariate(1.0 / 5.0) # 3activate(c, c.visit(timeInBank), at=t) # 4simulate(until=maxTime)

The result is shown below. The customer now arrives at about 0.64195, (or 10.58092 if you are not using Python 3).Changing the seed value would change that time.


The display looks pretty untidy. In the next example I will try and make it tidier.

If you are not using Python 3, your output may differ. The output for Python 2, for all examples, is given in AppendixA.

3.1.5 More customers

Our simulation does little so far. To consider a simulation with several customers we return to the simple deterministicmodel and add more Customers.

The program is almost as easy as the first example (A Customer arriving at a fixed time). The main change is between#4 to #5 where we create, name, and activate three customers. We also increase the maximum simulation time to 400(at #3 and referred to at #6). Observe that we need only one definition of the Customer class and create severalobjects of that class. These will act quite independently in this model.

Each customer stays for a different timeinbank so, instead of setting a common value for this we set it for eachcustomer. The customers are started at different times (using at=). Tony's activation time occurs before Klaus's,

3.1. The Bank 239


so Tony will arrive first even though his activation statement appears later in the script.

As promised, the print statements have been changed to use Python string formatting (at #1 and #2). The statementslook complicated but the output is much nicer.

""" bank02: More Customers """from SimPy.Simulation import *


class Customer(Process):""" Customer arrives, looks around and leaves """

def visit(self, timeInBank):print("%7.4f %s: Here I am" % (now(), self.name)) # 1yield hold, self, timeInBankprint("%7.4f %s: I must leave" % (now(), self.name)) # 2

# Experiment data -------------------------

maxTime = 400.0 # minutes #3


initialize()

c1 = Customer(name="Klaus") # 4activate(c1, c1.visit(timeInBank=10.0), at=5.0)c2 = Customer(name="Tony")activate(c2, c2.visit(timeInBank=7.0), at=2.0)c3 = Customer(name="Evelyn")activate(c3, c3.visit(timeInBank=20.0), at=12.0) # 5

simulate(until=maxTime) # 6

The trace produced by the program is shown below. Again the simulation finishes before the 400.0 specified in thesimulate call as it has run out of events.

2.0000 Tony: Here I am5.0000 Klaus: Here I am9.0000 Tony: I must leave

12.0000 Evelyn: Here I am15.0000 Klaus: I must leave32.0000 Evelyn: I must leave

3.1.6 Many customers

Another change will allow us to have more customers. As it is tedious to give a specially chosen name to each one, wewill call them Customer00, Customer01,... and use a separate Source class to create and activate them.To make things clearer we do not use the random numbers in this model.

The following listing shows the new program. At #1 to #4 a Source class is defined. Its PEM, here calledgenerate, is defined between #2 to #4. This PEM has a couple of arguments: the number of customers to begenerated and the Time Between Arrivals, TBA. It consists of a loop that creates a sequence of numbered Customersfrom 0 to (number-1), inclusive. We create a customer and give it a name at #3. It is then activated at the current



simulation time (the final argument of the activate statement is missing so that the default value of now() is usedas the time). We also specify how long the customer is to stay in the bank. To keep it simple, all customers stay exactly12 minutes. After each new customer is activated, the Source holds for a fixed time (yield hold,self,TBA)before creating the next one (at #4).

A Source, s, is created at #5 and activated at #6 where the number of customers to be generated is set to maxNumber= 5 and the interval between customers to ARRint = 10.0. Once started at time 0.0 it creates customers atintervals and each customer then operates independently of the others:

""" bank03: Many non-random Customers """from SimPy.Simulation import *


class Source(Process): # 1""" Source generates customers regularly """

def generate(self, number, TBA): # 2for i in range(number):

c = Customer(name="Customer%02d" % (i)) # 3activate(c, c.visit(timeInBank=12.0))yield hold, self, TBA # 4


def visit(self, timeInBank):print("%7.4f %s: Here I am" % (now(), self.name))yield hold, self, timeInBankprint("%7.4f %s: I must leave" % (now(), self.name))

# Experiment data -------------------------

maxNumber = 5maxTime = 400.0 # minutesARRint = 10.0 # time between arrivals, minutes


initialize()s = Source() # 5activate(s, s.generate(number=maxNumber, # 6

TBA=ARRint), at=0.0)simulate(until=maxTime)

The output is:

0.0000 Customer00: Here I am10.0000 Customer01: Here I am12.0000 Customer00: I must leave20.0000 Customer02: Here I am22.0000 Customer01: I must leave30.0000 Customer03: Here I am32.0000 Customer02: I must leave40.0000 Customer04: Here I am42.0000 Customer03: I must leave

3.1. The Bank 241


52.0000 Customer04: I must leave

3.1.7 Many random customers

We now extend this model to allow arrivals at random. In simulation this is usually interpreted as meaning that thetimes between customer arrivals are distributed as exponential random variates. There is little change in our program,we use a Source object, as before.

The exponential random variate is generated at #1 with meanTBA as the mean Time Between Arrivals and used at #2.Note that this parameter is not exactly intuitive. As already mentioned, the Python expovariate method uses therate of arrivals as the parameter not the average interval between them. The exponential delay between two arrivalsgives pseudo-random arrivals. In this model the first customer arrives at time 0.0.

The seed method is called to initialize the random number stream in the model routine (at #3). It is possible toleave this call out but if we wish to do serious comparisons of systems, we must have control over the random variatesand therefore control over the seeds. Then we can run identical models with different seeds or different models withidentical seeds. We provide the seeds as control parameters of the run. Here a seed is assigned at #3 but it is clear itcould have been read in or manually entered on an input form.

""" bank06: Many Random Customers """from SimPy.Simulation import *from random import expovariate, seed


class Source(Process):""" Source generates customers at random """

def generate(self, number, meanTBA):for i in range(number):

c = Customer(name="Customer%02d" % (i))activate(c, c.visit(timeInBank=12.0))t = expovariate(1.0 / meanTBA) # 1yield hold, self, t # 2


def visit(self, timeInBank=0):print("%7.4f %s: Here I am" % (now(), self.name))yield hold, self, timeInBankprint("%7.4f %s: I must leave" % (now(), self.name))

# Experiment data -------------------------

maxNumber = 5maxTime = 400.0 # minutesARRint = 10.0 # mean arrival interval, minutes


seed(99999) # 3initialize()



s = Source(name='Source')activate(s, s.generate(number=maxNumber,

meanTBA=ARRint), at=0.0)simulate(until=maxTime)

with the following output:

0.0000 Customer00: Here I am1.2839 Customer01: Here I am4.9842 Customer02: Here I am

12.0000 Customer00: I must leave13.2839 Customer01: I must leave16.9842 Customer02: I must leave35.5432 Customer03: Here I am47.5432 Customer03: I must leave48.9918 Customer04: Here I am60.9918 Customer04: I must leave

3.1. The Bank 243


3.2 Service Counters

We introduce a service counter at the Bank using a SimPy Resource.

3.2.1 A service counter

So far, the model has been more like an art gallery, the customers entering, looking around, and leaving. Now they aregoing to require service from the bank clerk. We extend the model to include a service counter which will be modelledas an object of SimPy’s Resource class with a single resource unit. The actions of a Resource are simple: acustomer requests a unit of the resource (a clerk). If one is free he gets service (and removes the unit). If there isno free clerk the customer joins the queue (managed by the resource object) until it is their turn to be served. As eachcustomer completes service and releases the unit, the clerk can start serving the next in line.

The service counter is created as a Resource (k) in at #8. This is provided as an argument to the Source (at #9)which, in turn, provides it to each customer it creates and activates (at #1).

The actions involving the service counter, k, in the customer’s PEM are:

• the yield request statement at #3. If the server is free then the customer can start service immediately andthe code moves on to #4. If the server is busy, the customer is automatically queued by the Resource. When iteventually comes available the PEM moves on to #4.

• the yield hold statement at #5 where the operation of the service counter is modelled. Here the service timeis a fixed timeInBank. During this period the customer is being served.

• the yield release statement at #6. The current customer completes service and the service counter be-comes available for any remaining customers in the queue.

Observe that the service counter is used with the pattern (yield request..; yield hold..; yieldrelease..).

To show the effect of the service counter on the activities of the customers, I have added #2 to record when the customerarrived and #4 to record the time between arrival in the bank and starting service. #4 is after the yield requestcommand and will be reached only when the request is satisfied. It is before the yield hold that corresponds to thestart of service. The variable wait will record how long the customer waited and will be 0 if he received service atonce. This technique of saving the arrival time in a variable is common. So the print statement also prints out howlong the customer waited in the bank before starting service.

""" bank07: One Counter,random arrivals """from SimPy.Simulation import *from random import expovariate, seed


class Source(Process):""" Source generates customers randomly """

def generate(self, number, meanTBA, resource):for i in range(number):

c = Customer(name="Customer%02d" % (i))activate(c, c.visit(timeInBank=12.0,

res=resource)) # 1t = expovariate(1.0 / meanTBA)yield hold, self, t

class Customer(Process):



""" Customer arrives, is served and leaves """

def visit(self, timeInBank, res):arrive = now() # 2 arrival timeprint("%8.3f %s: Here I am" % (now(), self.name))

yield request, self, res # 3wait = now() - arrive # 4 waiting timeprint("%8.3f %s: Waited %6.3f" % (now(), self.name, wait))yield hold, self, timeInBank # 5yield release, self, res # 6

print("%8.3f %s: Finished" % (now(), self.name))

# Experiment data -------------------------

maxNumber = 5 # 7maxTime = 400.0 # minutesARRint = 10.0 # mean, minutesk = Resource(name="Counter", unitName="Clerk") # 8

# Model/Experiment ------------------------------seed(99999)initialize()s = Source('Source')activate(s, s.generate(number=maxNumber,

meanTBA=ARRint, resource=k), at=0.0) # 9simulate(until=maxTime)

Examining the trace we see that the first, and last, customers get instant service but the others have to wait. We stillonly have five customers (#7) so we cannot draw general conclusions.

0.000 Customer00: Here I am0.000 Customer00: Waited 0.0001.284 Customer01: Here I am4.984 Customer02: Here I am12.000 Customer00: Finished12.000 Customer01: Waited 10.71624.000 Customer01: Finished24.000 Customer02: Waited 19.01635.543 Customer03: Here I am36.000 Customer02: Finished36.000 Customer03: Waited 0.45748.000 Customer03: Finished48.992 Customer04: Here I am48.992 Customer04: Waited 0.00060.992 Customer04: Finished

3.2.2 A server with a random service time

This is a simple change to the model in that we retain the single service counter but make the customer service timea random variable. As is traditional in the study of simple queues we first assume an exponential service time and setthe mean to timeInBank.

The service time random variable, tib, is generated at #1 and used at #2. The argument to be used in the call of

3.2. Service Counters 245


expovariate is not the mean of the distribution, timeInBank, but is the rate 1/timeInBank.

We have also collected together a number of constants by defining a number of appropriate variables and giving themvalues. These are in lines between marks #3 and #4.

""" bank08: A counter with a random service time """from SimPy.Simulation import *from random import expovariate, seed




c = Customer(name="Customer%02d" % (i,))activate(c, c.visit(b=resource))t = expovariate(1.0 / meanTBA)yield hold, self, t


def visit(self, b):arrive = now()print("%8.4f %s: Here I am " % (now(), self.name))yield request, self, bwait = now() - arriveprint("%8.4f %s: Waited %6.3f" % (now(), self.name, wait))tib = expovariate(1.0 / timeInBank) # 1yield hold, self, tib # 2yield release, self, bprint("%8.4f %s: Finished " % (now(), self.name))

# Experiment data -------------------------

maxNumber = 5 # 3maxTime = 400.0 # minutestimeInBank = 12.0 # mean, minutesARRint = 10.0 # mean, minutestheseed = 99999 # 4


seed(theseed)k = Resource(name="Counter", unitName="Clerk")

initialize()s = Source('Source')activate(s, s.generate(number=maxNumber, meanTBA=ARRint,

resource=k), at=0.0)simulate(until=maxTime)

And the output:



0.0000 Customer00: Here I am0.0000 Customer00: Waited 0.0001.2839 Customer01: Here I am4.4403 Customer00: Finished4.4403 Customer01: Waited 3.156

20.5786 Customer01: Finished31.8430 Customer02: Here I am31.8430 Customer02: Waited 0.00034.5594 Customer02: Finished36.2308 Customer03: Here I am36.2308 Customer03: Waited 0.00041.4313 Customer04: Here I am67.1315 Customer03: Finished67.1315 Customer04: Waited 25.70087.9241 Customer04: Finished

This model with random arrivals and exponential service times is an example of an M/M/1 queue and could rathereasily be solved analytically to calculate the steady-state mean waiting time and other operating characteristics. (Butnot so easily solved for its transient behavior.)

3.2.3 Several service counters

When we introduce several counters we must decide on a queue discipline. Are customers going to make one queue orare they going to form separate queues in front of each counter? Then there are complications - will they be allowedto switch lines (jockey)? We first consider a single queue with several counters and later consider separate isolatedqueues. We will not look at jockeying.

Here we model a bank whose customers arrive randomly and are to be served at a group of counters, taking a randomtime for service, where we assume that waiting customers form a single first-in first-out queue.

The only difference between this model and the single-server model is at #1. We have provided two counters byincreasing the capacity of the counter resource to 2. These units of the resource correspond to the two counters.Because both clerks cannot be called Karen, we have used a general name of Clerk.

""" bank09: Several Counters but a Single Queue """from SimPy.Simulation import *from random import expovariate, seed




c = Customer(name="Customer%02d" % (i))activate(c, c.visit(b=resource))t = expovariate(1.0 / meanTBA)yield hold, self, t


def visit(self, b):arrive = now()



print("%8.4f %s: Here I am " % (now(), self.name))yield request, self, bwait = now() - arriveprint("%8.4f %s: Waited %6.3f" % (now(), self.name, wait))tib = expovariate(1.0 / timeInBank)yield hold, self, tibyield release, self, bprint("%8.4f %s: Finished " % (now(), self.name))

# Experiment data -------------------------

maxNumber = 5maxTime = 400.0 # minutestimeInBank = 12.0 # mean, minutesARRint = 10.0 # mean, minutestheseed = 99999


seed(theseed)k = Resource(capacity=2, name="Counter", unitName="Clerk") # 1

initialize()s = Source('Source')activate(s, s.generate(number=maxNumber, meanTBA=ARRint,

resource=k), at=0.0)simulate(until=maxTime)

The waiting times in this model are very different from those for the single service counter. For example, none of thecustomers had to wait. But, again, we have observed too few customers to draw general conclusions.

0.0000 Customer00: Here I am0.0000 Customer00: Waited 0.0001.2839 Customer01: Here I am1.2839 Customer01: Waited 0.0004.4403 Customer00: Finished

17.4222 Customer01: Finished31.8430 Customer02: Here I am31.8430 Customer02: Waited 0.00034.5594 Customer02: Finished36.2308 Customer03: Here I am36.2308 Customer03: Waited 0.00041.4313 Customer04: Here I am41.4313 Customer04: Waited 0.00062.2239 Customer04: Finished67.1315 Customer03: Finished

3.2.4 Several counters with individual queues

Each counter now has its own queue. The programming is more complicated because the customer has to decidewhich one to join. The obvious technique is to make each counter a separate resource and it is useful to make a list ofresource objects (#7).

In practice, a customer will join the shortest queue. So we define a Python function, NoInSystem(R) (#1 to #2) toreturn the sum of the number waiting and the number being served for a particular counter, R. This function is used



at #3 to list the numbers at each counter. It is then easy to find which counter the arriving customer should join. Wehave also modified the trace printout, #4 to display the state of the system when the customer arrives. We choose theshortest queue at #5 to #6 (using the variable choice).

The rest of the program is the same as before.

""" bank10: Several Counters with individual queues"""from SimPy.Simulation import *from random import expovariate, seed



def generate(self, number, interval, counters):for i in range(number):

c = Customer(name="Customer%02d" % (i))activate(c, c.visit(counters))t = expovariate(1.0 / interval)yield hold, self, t

def NoInSystem(R): # 1""" Total number of customers in the resource R"""return (len(R.waitQ) + len(R.activeQ)) # 2

class Customer(Process):""" Customer arrives, chooses the shortest queue

is served and leaves"""

def visit(self, counters):arrive = now()Qlength = [NoInSystem(counters[i]) for i in range(Nc)] # 3print("%7.4f %s: Here I am. %s" % (now(), self.name, Qlength)) # 4for i in range(Nc): # 5

if Qlength[i] == 0 or Qlength[i] == min(Qlength):choice = i # the index of the shortest linebreak # 6

yield request, self, counters[choice]wait = now() - arriveprint("%7.4f %s: Waited %6.3f" % (now(), self.name, wait))tib = expovariate(1.0 / timeInBank)yield hold, self, tibyield release, self, counters[choice]

print("%7.4f %s: Finished" % (now(), self.name))

# Experiment data -------------------------

maxNumber = 5maxTime = 400.0 # minutestimeInBank = 12.0 # mean, minutesARRint = 10.0 # mean, minutes



Nc = 2 # number of counterstheseed = 787878


seed(theseed)kk = [Resource(name="Clerk0"), Resource(name="Clerk1")] # 7initialize()s = Source('Source')activate(s, s.generate(number=maxNumber, interval=ARRint,

counters=kk), at=0.0)simulate(until=maxTime)

The results show how the customers choose the counter with the smallest number. Unlucky Customer03 who joinsthe wrong queue has to wait until Customer01 finishes before his service can be started. There are, however, toofew arrivals in these runs, limited as they are to five customers, to draw any general conclusions about the relativeefficiencies of the two systems.

0.0000 Customer00: Here I am. [0, 0]0.0000 Customer00: Waited 0.0009.7519 Customer00: Finished

12.0829 Customer01: Here I am. [0, 0]12.0829 Customer01: Waited 0.00025.9167 Customer02: Here I am. [1, 0]25.9167 Customer02: Waited 0.00038.2349 Customer03: Here I am. [1, 1]40.4032 Customer04: Here I am. [2, 1]43.0677 Customer02: Finished43.0677 Customer04: Waited 2.66444.0242 Customer01: Finished44.0242 Customer03: Waited 5.78960.1271 Customer03: Finished70.2500 Customer04: Finished



3.3 Customer’s Priority

In Many situations there is a system of priority service. Those customers with high priority are served first, those withlow priority must wait. In some cases, preemptive priority will even allow a high-priority customer to interrupt theservice of one with a lower priority.

3.3.1 Priority customers

SimPy implements priority requests with an extra numerical priority argument in the yield request command,higher values meaning higher priority. For this to operate, the requested Resource must have been defined withqType=PriorityQ.

In the first example, we modify the program with random arrivals, one counter, and a fixed service time with theaddition of a high priority customer. Warning: The seed() value has been changed to 787878 to make the storymore exciting. To make things even more confusing, your results may be different from those here because therandom module gives different results for Python 2.x and 3.x.,

The main modifications are to the definition of the counter where we change the qType and to the yieldrequest command in the visit PEM of the customer. We must provide each customer with a priority. Sincethe default is priority=0 this is easy for most of them.

To observe the priority in action, while all other customers have the default priority of 0, at between #5 and #6 wecreate and activate one special customer, Guido, with priority 100 who arrives at time 23.0.

The visit customer method has a new parameter, P (at #3), which allows us to set the customer priority.

At #4, counter is defined with qType=PriorityQ so that we can request it with priority (at #3) using thestatement yield request,self,counter,P

At #2, we now print out the number of customers waiting when each customer arrives.

""" bank20: One counter with a priority customer """from SimPy.Simulation import *from random import expovariate, seed



def generate(self, number, interval, resource):for i in range(number):


res=resource, P=0))t = expovariate(1.0 / interval)yield hold, self, t


def visit(self, timeInBank=0, res=None, P=0): # 1arrive = now() # arrival timeNwaiting = len(res.waitQ)print("%8.3f %s: Queue is %d on arrival" % # 2

(now(), self.name, Nwaiting))

3.3. Customer’s Priority 251


yield request, self, res, P # 3wait = now() - arrive # waiting timeprint("%8.3f %s: Waited %6.3f" %

(now(), self.name, wait))yield hold, self, timeInBankyield release, self, res

print("%8.3f %s: Completed" %(now(), self.name))

# Experiment data -------------------------

maxTime = 400.0 # minutesk = Resource(name="Counter", unitName="Karen", # 4

qType=PriorityQ)

# Model/Experiment ------------------------------seed(787878)initialize()s = Source('Source')activate(s, s.generate(number=5, interval=10.0,

resource=k), at=0.0)guido = Customer(name="Guido ") # 5activate(guido, guido.visit(timeInBank=12.0, res=k, # 6

P=100), at=23.0)simulate(until=maxTime)

The resulting output is as follows. The number of customers in the queue just as each arrives is displayed in the trace.That count does not include any customer in service.

0.000 Customer00: Queue is 0 on arrival0.000 Customer00: Waited 0.00012.000 Customer00: Completed12.083 Customer01: Queue is 0 on arrival12.083 Customer01: Waited 0.00020.210 Customer02: Queue is 0 on arrival23.000 Guido : Queue is 1 on arrival24.083 Customer01: Completed24.083 Guido : Waited 1.08334.043 Customer03: Queue is 1 on arrival36.083 Guido : Completed36.083 Customer02: Waited 15.87348.083 Customer02: Completed48.083 Customer03: Waited 14.04060.083 Customer03: Completed60.661 Customer04: Queue is 0 on arrival60.661 Customer04: Waited 0.00072.661 Customer04: Completed

Reading carefully one can see that when Guido arrives Customer00 has been served and left at 12.000,Customer01 is in service and Customer02 is waiting in the queue. Guido has priority over any waiting cus-tomers and is served before the at 24.083. When Guido leaves at 36.083, Customer02 starts service havingwaited 15.873 minutes.



3.3.2 A priority customer with preemption

Now we allow Guido to have preemptive priority. He will displace any customer in service when he arrives. Thatcustomer will resume when Guido finishes (unless higher priority customers intervene). It requires only a change toone line of the program, adding the argument, preemptable=True to the Resource statement at #1.

""" bank23: One counter with a priority customer with preemption """from SimPy.Simulation import *from random import expovariate, seed





res=resource, P=0))t = expovariate(1.0 / interval)yield hold, self, t


def visit(self, timeInBank=0, res=None, P=0):arrive = now() # arrival timeNwaiting = len(res.waitQ)print("%8.3f %s: Queue is %d on arrival" %

(now(), self.name, Nwaiting))

yield request, self, res, Pwait = now() - arrive # waiting timeprint("%8.3f %s: Waited %6.3f" %

(now(), self.name, wait))yield hold, self, timeInBankyield release, self, res

print("%8.3f %s: Completed" %(now(), self.name))

# Experiment data -------------------------

maxTime = 400.0 # minutesk = Resource(name="Counter", unitName="Karen", # 1

qType=PriorityQ, preemptable=True)

# Model/Experiment ------------------------------seed(989898)initialize()s = Source('Source')activate(s, s.generate(number=5, interval=10.0,

resource=k), at=0.0)guido = Customer(name="Guido ")

3.3. Customer’s Priority 253


activate(guido, guido.visit(timeInBank=12.0, res=k,P=100), at=23.0)


Though Guido arrives at the same time, 23.000, he no longer has to wait and immediately goes into service,displacing the incumbent, Customer01. That customer had already completed 23.000-12.083 = 10.917minutes of his service. When Guido finishes at 36.083, Customer01 resumes service and takes 36.083-35.000 = 1.083 minutes to finish. His total service time is the same as before (12.000 minutes).

0.000 Customer00: Queue is 0 on arrival0.000 Customer00: Waited 0.0008.634 Customer01: Queue is 0 on arrival12.000 Customer00: Completed12.000 Customer01: Waited 3.36616.016 Customer02: Queue is 0 on arrival19.882 Customer03: Queue is 1 on arrival20.246 Customer04: Queue is 2 on arrival23.000 Guido : Queue is 3 on arrival23.000 Guido : Waited 0.00035.000 Guido : Completed36.000 Customer01: Completed36.000 Customer02: Waited 19.98448.000 Customer02: Completed48.000 Customer03: Waited 28.11860.000 Customer03: Completed60.000 Customer04: Waited 39.75472.000 Customer04: Completed



3.4 Balking and Reneging Customers

Balking occurs when a customer refuses to join a queue if it is too long. Reneging (or, better, abandonment) occurs ifan impatient customer gives up while still waiting and before being served.

3.4.1 Balking customers

Another term for a system with balking customers is one where “blocked customers” are “cleared”, termed by en-gineers a BCC system. This is very convenient analytically in queueing theory and formulae developed using thisassumption are used extensively for planning communication systems. The easiest case is when no queueing is al-lowed.

As an example let us investigate a BCC system with a single server but the waiting space is limited. We will estimatethe rate of balking when the maximum number in the queue is set to 1. On arrival into the system the customer mustfirst check to see if there is room. We will need the number of customers in the system or waiting. We could keep acount, incrementing when a customer joins the queue or, since we have a Resource, use the length of the Resource’swaitQ. Choosing the latter we test (at #1). If there is not enough room, we balk, incrementing a Class variableCustomer.numBalking at #2 to get the total number balking during the run.

""" bank24. BCC system with several counters """from SimPy.Simulation import *from random import expovariate, seed




c = Customer(name="Customer%02d" % (i))activate(c, c.visit(b=resource))t = expovariate(1.0 / meanTBA)yield hold, self, t


numBalking = 0

def visit(self, b):arrive = now()print("%8.4f %s: Here I am " %

(now(), self.name))if len(b.waitQ) < maxInQueue: # 1

yield request, self, bwait = now() - arriveprint("%8.4f %s: Wait %6.3f" %

(now(), self.name, wait))tib = expovariate(1.0 / timeInBank)yield hold, self, tibyield release, self, bprint("%8.4f %s: Finished " %

(now(), self.name))

3.4. Balking and Reneging Customers 255


else:Customer.numBalking += 1 # 2print("%8.4f %s: BALKING " %

(now(), self.name))

# Experiment data -------------------------------

timeInBank = 12.0 # mean, minutesARRint = 10.0 # mean interarrival time, minutesnumServers = 1 # serversmaxInSystem = 2 # customersmaxInQueue = maxInSystem - numServers

maxNumber = 8maxTime = 4000.0 # minutestheseed = 212121


seed(theseed)k = Resource(capacity=numServers,

name="Counter", unitName="Clerk")

initialize()s = Source('Source')activate(s, s.generate(number=maxNumber,

meanTBA=ARRint,resource=k), at=0.0)


# Results -----------------------------------------

nb = float(Customer.numBalking)print("balking rate is %8.4f per minute" % (nb / now()))

The resulting output for a run of this program showing balking occurring is given below:

0.0000 Customer00: Here I am0.0000 Customer00: Wait 0.0004.3077 Customer01: Here I am5.6957 Customer02: Here I am5.6957 Customer02: BALKING6.9774 Customer03: Here I am6.9774 Customer03: BALKING8.2476 Customer00: Finished8.2476 Customer01: Wait 3.940

21.1312 Customer04: Here I am22.4840 Customer01: Finished22.4840 Customer04: Wait 1.35323.0923 Customer05: Here I am23.1537 Customer06: Here I am23.1537 Customer06: BALKING36.0653 Customer04: Finished36.0653 Customer05: Wait 12.97338.4851 Customer07: Here I am53.1056 Customer05: Finished53.1056 Customer07: Wait 14.620



60.3558 Customer07: Finishedbalking rate is 0.0497 per minute

When Customer02 arrives, Customer00 is already in service and Customer01 is waiting. There is no room soCustomer02 balks. In fact another customer, Customer03 arrives and balks before Customer00 is finished.

The balking pattern for python 2.x is different.

3.4.2 Reneging or abandoning

Often in practice an impatient customer will leave the queue before being served. SimPy can model this renegingbehaviour using a compound yield statement. In such a statement there are two yield clauses. An example is:

yield (request,self,counter),(hold,self,maxWaitTime)

The first tuple of this statement is the usual yield request, asking for a unit of counter Resource. The processwill either get the unit immediately or be queued by the Resource. The second tuple is a reneging clause which hasthe same syntax as a yield hold. The requesting process will renege if the wait exceeds maxWaitTime.

There is a complication, though. The requesting PEM must discover what actually happened. Did the process get theresource or did it renege? This involves a mandatory test of self.acquired(resource). In our example, this testis at #1.

""" bank21: One counter with impatient customers """from SimPy.Simulation import *from random import expovariate, seed



def generate(self, number, interval):for i in range(number):

c = Customer(name="Customer%02d" % (i))activate(c, c.visit(timeInBank=15.0))t = expovariate(1.0 / interval)yield hold, self, t


def visit(self, timeInBank=0):arrive = now() # arrival timeprint("%8.3f %s: Here I am " % (now(), self.name))

yield (request, self, counter), (hold, self, maxWaitTime)wait = now() - arrive # waiting timeif self.acquired(counter): # 1

print("%8.3f %s: Waited %6.3f" % (now(), self.name, wait))yield hold, self, timeInBankyield release, self, counterprint("%8.3f %s: Completed" % (now(), self.name))

else:print("%8.3f %s: Waited %6.3f. I am off" %

3.4. Balking and Reneging Customers 257


(now(), self.name, wait))

# Experiment data -------------------------

maxTime = 400.0 # minutesmaxWaitTime = 12.0 # minutes. maximum time to wait

# Model ----------------------------------

def model():global counterseed(989898)counter = Resource(name="Karen")initialize()source = Source('Source')activate(source,

source.generate(number=5, interval=10.0), at=0.0)simulate(until=maxTime)

# Experiment ----------------------------------

model()

0.000 Customer00: Here I am0.000 Customer00: Waited 0.0008.634 Customer01: Here I am15.000 Customer00: Completed15.000 Customer01: Waited 6.36616.016 Customer02: Here I am19.882 Customer03: Here I am20.246 Customer04: Here I am28.016 Customer02: Waited 12.000. I am off30.000 Customer01: Completed30.000 Customer03: Waited 10.11832.246 Customer04: Waited 12.000. I am off45.000 Customer03: Completed

Customer02 arrives at 16.016 but has only 12 minutes patience. After that time in the queue (at time 28.016) heabandons the queue to leave 03 to take his place. 04 also abandons.

The reneging pattern for python 2.x is different due to a different expovariate implementation.

3.5 Gathering Statistics

SimPy Monitors allow statistics to be gathered and simple summaries calculated.

3.5.1 The bank with a monitor

The traces of output that have been displayed so far are valuable for checking that the simulation is operating correctlybut would become too much if we simulate a whole day. We do need to get results from our simulation to answer theoriginal questions. What, then, is the best way to summarize the results?



One way is to analyze the traces elsewhere, piping the trace output, or a modified version of it, into a real statisticalprogram such as R for statistical analysis, or into a file for later examination by a spreadsheet. We do not have spaceto examine this thoroughly here. Another way of presenting the results is to provide graphical output.

SimPy offers an easy way to gather a few simple statistics such as averages: the Monitor and Tally classes. TheMonitor records the values of chosen variables as time series (but see the comments in Final Remarks).

We now demonstrate a Monitor that records the average waiting times for our customers. We return to the systemwith random arrivals, random service times and a single queue and remove the old trace statements. In practice, wewould make the printouts controlled by a variable, say, TRACE which is set in the experimental data (or read in as aprogram option - but that is a different story). This would aid in debugging and would not complicate the data analysis.We will run the simulations for many more arrivals.

A Monitor, wM, is created at #2. It observes and records the waiting time mentioned at #1. We run maxNumber=50customers (in the call of generate at #3) and have increased maxTime to 1000 minutes. Brief statistics are givenby the Monitor methods count() and mean() at #4.

""" bank11: The bank with a Monitor"""from SimPy.Simulation import *from random import expovariate, seed




c = Customer(name="Customer%02d" % (i))activate(c, c.visit(b=resource))t = expovariate(1.0 / interval)yield hold, self, t


def visit(self, b):arrive = now()yield request, self, bwait = now() - arrivewM.observe(wait) # 1tib = expovariate(1.0 / timeInBank)yield hold, self, tibyield release, self, b

# Experiment data -------------------------

maxNumber = 50maxTime = 1000.0 # minutestimeInBank = 12.0 # mean, minutesARRint = 10.0 # mean, minutesNc = 2 # number of counterstheseed = 99999

# Model/Experiment ----------------------

3.5. Gathering Statistics 259


seed(theseed)k = Resource(capacity=Nc, name="Clerk")wM = Monitor() # 2initialize()s = Source('Source')activate(s, s.generate(number=maxNumber, interval=ARRint,

resource=k), at=0.0) # 3simulate(until=maxTime)

# Result ----------------------------------

result = wM.count(), wM.mean() # 4print("Average wait for %3d completions was %5.3f minutes." % result)

The average waiting time for 50 customers in this 2-counter system is more reliable (i.e., less subject to randomsimulation effects) than the times we measured before but it is still not sufficiently reliable for real-world decisions.We should also replicate the runs using different random number seeds. The result of this run (using Python 3.2) is:

Average wait for 50 completions was 8.941 minutes.

Result for Python 2.x is given in Appendix A.

3.5.2 Monitoring a resource

Now consider observing the number of customers waiting or executing in a Resource. Because of the need toobserve the value after the change but at the same simulation instant, it is impossible to use the length of theResource’s waitQ directly with a Monitor defined outside the Resource. Instead Resources can be set up with built-inMonitors.

Here is an example using a Monitored Resource. We intend to observe the average number waiting and active in thecounter resource. counter is defined at #1 and we have set monitored=True. This establishes two Monitors:waitMon, to record changes in the numbers waiting and actMon to record changes in the numbers active in thecounter. We need make no further change to the operation of the program as monitoring is then automatic. Noobserve calls are necessary.

At the end of the run in the model function, we calculate the timeAverage of both waitMon and actMon andreturn them from the model call (at #2). These can then be printed at the end of the program (at #3).

"""bank15: Monitoring a Resource"""from SimPy.Simulation import *from random import expovariate, seed



def generate(self, number, rate):for i in range(number):

c = Customer(name="Customer%02d" % (i))activate(c, c.visit(timeInBank=12.0))yield hold, self, expovariate(rate)

class Customer(Process):



""" Customer arrives, is served and leaves """

def visit(self, timeInBank):arrive = now()print("%8.4f %s: Arrived " % (now(), self.name))

yield request, self, counterprint("%8.4f %s: Got counter " % (now(), self.name))tib = expovariate(1.0 / timeInBank)yield hold, self, tibyield release, self, counter

print("%8.4f %s: Finished " % (now(), self.name))

# Experiment data -------------------------maxTime = 400.0 # minutescounter = Resource(1, name="Clerk", monitored=True) # 1

# Model ----------------------------------

def model(SEED=393939):seed(SEED)initialize()source = Source()activate(source,

source.generate(number=5, rate=0.1), at=0.0)simulate(until=maxTime)

return (counter.waitMon.timeAverage(),counter.actMon.timeAverage()) # 2

# Experiment ----------------------------------

print('Avge waiting = %6.4f\nAvge active = %6.4f\n' % model()) # 3

With the following output:

0.0000 Customer00: Arrived0.0000 Customer00: Got counter1.1489 Customer01: Arrived5.6657 Customer02: Arrived

10.0126 Customer00: Finished10.0126 Customer01: Got counter12.8923 Customer03: Arrived18.5883 Customer04: Arrived29.3088 Customer01: Finished29.3088 Customer02: Got counter44.1219 Customer02: Finished44.1219 Customer03: Got counter73.0066 Customer03: Finished73.0066 Customer04: Got counter73.8458 Customer04: Finished

Avge waiting = 1.6000Avge active = 1.0000

3.5. Gathering Statistics 261


3.5.3 Multiple runs

To get a number of independent measurements we must replicate the runs using different random number seeds. Eachreplication must be independent of previous ones so the Monitor and Resources must be redefined for each run. Wecan no longer allow them to be global objects as we have before.

We will define a function, model with a parameter runSeed so that the random number seed can be different fordifferent runs (between between #2 and #5). The contents of the function are the same as the Model/Experimentin bank11: The bank with a monitor, except for one vital change.

This is required since the Monitor, wM, is defined inside the model function (#3). A customer can no longer refer toit. In the spirit of quality computer programming we will pass wM as a function argument. Unfortunately we have todo this in two steps, first to the Source (#4) and then from the Source to the Customer (#1).

model() is run for four different random-number seeds to get a set of replications (between #6 and #7).

""" bank12: Multiple runs of the bank with a Monitor"""from SimPy.Simulation import *from random import expovariate, seed



def generate(self, number, interval, resource, mon):for i in range(number):

c = Customer(name="Customer%02d" % (i))activate(c, c.visit(b=resource, M=mon)) # 1t = expovariate(1.0 / interval)yield hold, self, t


def visit(self, b, M):arrive = now()yield request, self, bwait = now() - arriveM.observe(wait)tib = expovariate(1.0 / timeInBank)yield hold, self, tibyield release, self, b

# Experiment data -------------------------

maxNumber = 50maxTime = 2000.0 # minutestimeInBank = 12.0 # mean, minutesARRint = 10.0 # mean, minutesNc = 2 # number of counterstheSeed = 393939

# Model ----------------------------------



def model(runSeed=theSeed): # 2seed(runSeed)k = Resource(capacity=Nc, name="Clerk")wM = Monitor() # 3

initialize()s = Source('Source')activate(s, s.generate(number=maxNumber, interval=ARRint,

resource=k, mon=wM), at=0.0) # 4simulate(until=maxTime)return (wM.count(), wM.mean()) # 5

# Experiment/Result ----------------------------------

theseeds = [393939, 31555999, 777999555, 319999771] # 6for Sd in theseeds:

result = model(Sd)print("Avge wait for %3d completions was %6.2f min." %

result) # 7

The results show some variation. Remember, though, that the system is still only operating for 50 customers so thesystem may not be in steady-state.

Avge wait for 50 completions was 3.66 min.Avge wait for 50 completions was 2.62 min.Avge wait for 50 completions was 8.97 min.Avge wait for 50 completions was 5.34 min.

3.6 Final Remarks

This introduction is too long and the examples are getting longer. There is much more to say about simulation withSimPy but no space. I finish with a list of topics for further study:

3.6.1 Topics not yet mentioned

• GUI input. Graphical input of simulation parameters could be an advantage in some cases. SimPy allows thisand programs using these facilities have been developed (see, for example, program MM1.py in the examplesin the SimPy distribution)

• Graphical Output. Similarly, graphical output of results can also be of value, not least in debugging simulationprograms and checking for steady-state conditions. SimPlot is useful here.

• Statistical Output. The Monitor class is useful in presenting results but more powerful methods of analysisare often needed. One solution is to output a trace and read that into a large-scale statistical system such as R.

• Priorities and Reneging in queues. SimPy allows processes to request units of resources under a priority queuediscipline (preemptive or not). It also allows processes to renege from a queue.

• Other forms of Resource Facilities. SimPy has two other resource structures: Levels to hold bulk commodi-ties, and Stores to contain an inventory of different object types.

• Advanced synchronization/scheduling commands. SimPy allows process synchronization by events and sig-nals.

3.6. Final Remarks 263


3.6.2 Acknowledgements

I thank Klaus Muller, Bob Helmbold, Mukhlis Matti, Karen Turner and other developers and users of SimPy forimproving this document by sending their comments. I would be grateful for further suggestions or corrections.Please send them to: vignaux at users.sourceforge.net.

3.6.3 References

• Python website: https://www.python.org

• SimPy Classic website: https://github.com/SimPyClassic/SimPyClassic

3.7 Appendix A

With Python 3 the definition of expovariate changed. In some cases this was back ported to some distributions ofPython 2.7. Because of this the output for the bank programs varies. This section contains the older output producedby the original definition of expovariate.

A Customer arriving at a fixed time


A Customer arriving at random


More Customers



Many Customers

0.0000 Customer00: Here I am10.0000 Customer01: Here I am12.0000 Customer00: I must leave20.0000 Customer02: Here I am22.0000 Customer01: I must leave30.0000 Customer03: Here I am32.0000 Customer02: I must leave40.0000 Customer04: Here I am42.0000 Customer03: I must leave52.0000 Customer04: I must leave

Many Random Customers

0.0000 Customer00: Here I am12.0000 Customer00: I must leave21.1618 Customer01: Here I am





32.8968 Customer02: Here I am33.1618 Customer01: I must leave33.3790 Customer03: Here I am36.3979 Customer04: Here I am44.8968 Customer02: I must leave45.3790 Customer03: I must leave48.3979 Customer04: I must leave

A Service Counter

0.000 Customer00: Here I am0.000 Customer00: Waited 0.00012.000 Customer00: Finished21.162 Customer01: Here I am21.162 Customer01: Waited 0.00032.897 Customer02: Here I am33.162 Customer01: Finished33.162 Customer02: Waited 0.26533.379 Customer03: Here I am36.398 Customer04: Here I am45.162 Customer02: Finished45.162 Customer03: Waited 11.78357.162 Customer03: Finished57.162 Customer04: Waited 20.76469.162 Customer04: Finished

A server with a random service time

0.0000 Customer00: Here I am0.0000 Customer00: Waited 0.000

14.0819 Customer00: Finished21.1618 Customer01: Here I am21.1618 Customer01: Waited 0.00021.6441 Customer02: Here I am24.7845 Customer01: Finished24.7845 Customer02: Waited 3.14031.9954 Customer03: Here I am41.0215 Customer04: Here I am43.9441 Customer02: Finished43.9441 Customer03: Waited 11.94949.9552 Customer03: Finished49.9552 Customer04: Waited 8.93452.2900 Customer04: Finished

Several Counters but a Single Queue

0.0000 Customer00: Here I am0.0000 Customer00: Waited 0.000

14.0819 Customer00: Finished21.1618 Customer01: Here I am21.1618 Customer01: Waited 0.00021.6441 Customer02: Here I am21.6441 Customer02: Waited 0.00024.7845 Customer01: Finished31.9954 Customer03: Here I am31.9954 Customer03: Waited 0.00032.9459 Customer03: Finished40.8037 Customer02: Finished

3.7. Appendix A 265


41.0215 Customer04: Here I am41.0215 Customer04: Waited 0.00043.3562 Customer04: Finished

Several Counters with individual queues

0.0000 Customer00: Here I am. [0, 0]0.0000 Customer00: Waited 0.0003.5483 Customer01: Here I am. [1, 0]3.5483 Customer01: Waited 0.0004.4169 Customer01: Finished6.4349 Customer02: Here I am. [1, 0]6.4349 Customer02: Waited 0.0007.0368 Customer00: Finished9.7200 Customer02: Finished9.8846 Customer03: Here I am. [0, 0]9.8846 Customer03: Waited 0.000

19.8340 Customer03: Finished26.2357 Customer04: Here I am. [0, 0]26.2357 Customer04: Waited 0.00029.8709 Customer04: Finished

Priority Customers

0.000 Customer00: Queue is 0 on arrival0.000 Customer00: Waited 0.0003.548 Customer01: Queue is 0 on arrival9.412 Customer02: Queue is 1 on arrival12.000 Customer00: Completed12.000 Customer01: Waited 8.45212.299 Customer03: Queue is 1 on arrival13.023 Customer04: Queue is 2 on arrival23.000 Guido : Queue is 3 on arrival24.000 Customer01: Completed24.000 Guido : Waited 1.00036.000 Guido : Completed36.000 Customer02: Waited 26.58848.000 Customer02: Completed48.000 Customer03: Waited 35.70160.000 Customer03: Completed60.000 Customer04: Waited 46.97772.000 Customer04: Completed

A priority Customer with Preemption

0.000 Customer00: Queue is 0 on arrival0.000 Customer00: Waited 0.0005.477 Customer01: Queue is 0 on arrival11.978 Customer02: Queue is 1 on arrival12.000 Customer00: Completed12.000 Customer01: Waited 6.52323.000 Guido : Queue is 1 on arrival23.000 Guido : Waited 0.00023.350 Customer03: Queue is 2 on arrival35.000 Guido : Completed36.000 Customer01: Completed36.000 Customer02: Waited 24.02248.000 Customer02: Completed



48.000 Customer03: Waited 24.65056.664 Customer04: Queue is 0 on arrival60.000 Customer03: Completed60.000 Customer04: Waited 3.33672.000 Customer04: Completed

Balking Customers

0.0000 Customer00: Here I am0.0000 Customer00: Wait 0.0008.3884 Customer00: Finished

10.4985 Customer01: Here I am10.4985 Customer01: Wait 0.00030.9314 Customer02: Here I am33.7131 Customer03: Here I am33.7131 Customer03: BALKING35.9122 Customer01: Finished35.9122 Customer02: Wait 4.98137.3562 Customer04: Here I am41.2491 Customer05: Here I am41.2491 Customer05: BALKING56.6185 Customer02: Finished56.6185 Customer04: Wait 19.26259.5365 Customer04: Finished92.2071 Customer06: Here I am92.2071 Customer06: Wait 0.00094.9740 Customer07: Here I am

102.7425 Customer06: Finished102.7425 Customer07: Wait 7.769133.5005 Customer07: Finishedbalking rate is 0.0150 per minute

Reneging or Abandoning

The Bank with a Monitor


Monitoring a Resource

0.0000 Customer00: Arrived0.0000 Customer00: Got counter6.8329 Customer00: Finished

22.2064 Customer01: Arrived22.2064 Customer01: Got counter30.1802 Customer01: Finished32.3277 Customer02: Arrived32.3277 Customer02: Got counter34.5627 Customer03: Arrived42.3374 Customer02: Finished42.3374 Customer03: Got counter46.4642 Customer03: Finished61.7697 Customer04: Arrived61.7697 Customer04: Got counter94.1098 Customer04: Finished

Avge waiting = 0.0826

3.7. Appendix A 267


Avge active = 0.6512

Multiple runs




3.8 Introduction

The first Bank tutorial, The Bank, developed and explained a series of simulation models of a simple bank usingSimPy. In various models, customers arrived randomly, queued up to be served at one or several counters, modelledusing the Resource class, and, in one case, could choose the shortest among several queues. It demonstrated the use ofthe Monitor class to record delays and showed how a model() mainline for the simulation was convenient to executereplications of simulation runs.

In this extension to The Bank, I provide more examples of SimPy facilities for which there was no room and forsome that were developed since it was written. These facilities are generally more complicated than those introducedbefore. They include queueing with priority, possibly with preemption, reneging, plotting, interrupting, waiting untila condition occurs (waituntil) and waiting for events to occur.

The programs are available without line numbers and ready to go, in directory bankprograms. Some have tracestatements for demonstration purposes, others produce graphical output to the screen. Let me encourage you to runthem and modify them for yourself

SimPy itself can be obtained from: https://github.com/SimPyClassic/SimPyClassic. It is compatible with Pythonversion 2.7 onwards. The examples in this documentation run with SimPy version 1.5 and later.

This tutorial should be read with the SimPy Manual or Cheatsheet at your side for reference.

3.8. Introduction 269




3.9 Processes

In some simulations it is valuable for one SimPy Process to interrupt another. This can only be done when the victimis “active”; that is when it has an event scheduled for it. It must be executing a yield hold statement.

A process waiting for a resource (after a yield request statement) is passive and cannot be interrupted by another.Instead the yield waituntil and yield waitevent facilities have been introduced to allow processes to waitfor conditions set by other processes.

3.9.1 Interrupting a Process.

Klaus goes into the bank to talk to the manager. For clarity we ignore the counters and other customers. During hisconversation his cellphone rings. When he finishes the call he continues the conversation.

In this example, call is an object of the Call Process class whose only purpose is to make the cellphone ring aftera delay, timeOfCall, an argument to its ring PEM (label #7).

klaus, a Customer, is interrupted by the call (Label #8). He is in the middle of a yield hold (label #1). Whenhe exits from that command it is as if he went into a trance when talking to the bank manager. He suddenly wakes upand must check (label #2) to see whether has finished his conversation (if there was no call) or has been interrupted.

If self.interrupted() is False he was not interrupted and leaves the bank (label #6) normally. If it is True,he was interrupted by the call, remembers how much conversation he has left (Label #3), resets the interrupt and thendeals with the call. When he finishes (label #4) he can resume the conversation, with, now we assume, a thoroughlyirritated bank manager (label #5).

""" bank22: An interruption by a phone call """from SimPy.Simulation import *



def visit(self, timeInBank, onphone):print("%7.4f %s: Here I am" % (now(), self.name))yield hold, self, timeInBank # 1if self.interrupted(): # 2

timeleft = self.interruptLeft # 3self.interruptReset()print("%7.4f %s: Excuse me" % (now(), self.name))print("%7.4f %s: Hello! I'll call back" % (now(), self.name))yield hold, self, onphoneprint("%7.4f %s: Sorry, where were we?" % (now(), self.name)) # 4yield hold, self, timeleft # 5

print("%7.4f %s: I must leave" % (now(), self.name)) # 6

class Call(Process):""" Cellphone call arrives and interrupts """

def ring(self, klaus, timeOfCall): # 7yield hold, self, timeOfCall # 8print("%7.4f Ringgg!" % (now()))self.interrupt(klaus)



# Experiment data -------------------------

timeInBank = 20.0timeOfCall = 9.0onphone = 3.0maxTime = 100.0

# Model/Experiment ----------------------------------

initialize()klaus = Customer(name="Klaus")activate(klaus, klaus.visit(timeInBank, onphone))call = Call(name="klaus")activate(call, call.ring(klaus, timeOfCall))simulate(until=maxTime)

0.0000 Klaus: Here I am9.0000 Ringgg!9.0000 Klaus: Excuse me9.0000 Klaus: Hello! I'll call back

12.0000 Klaus: Sorry, where were we?23.0000 Klaus: I must leave

As this has no random numbers the results are reasonably clear: the interrupting call occurs at 9.0. It takes klaus3 minutes to listen to the message and he resumes the conversation with the bank manager at 12.0. His total time ofconversation is 9.0 + 11.0 = 20.0 minutes as it would have been if the interrupt had not occurred.

3.9.2 waituntil the Bank door opens

Customers arrive at random, some of them getting to the bank before the door is opened by a doorman. They wait forthe door to be opened and then rush in and queue to be served.

This model uses the waituntil yield command. In the program listing the door is initially closed (label #1) and afunction to test if it is open is defined at label #2.

The Doorman class is defined starting at label #3 and the single doorman is created and activated at at labels #7 and#8. The doorman waits for an average 10 minutes (label #4) and then opens the door.

The Customer class is defined at label #5 and a new customer prints out Here I am on arrival. If the door is stillclosed, he adds but the door is shut and settles down to wait (label #6), using the yield waituntilcommand. When the door is opened by the doorman the dooropen state is changed and the customer (and all otherswaiting for the door) proceed. A customer arriving when the door is open will not be delayed.

"""bank14: *waituntil* the Bank door opens"""from SimPy.Simulation import *from random import expovariate, seed

# Model components ------------------------door = 'Shut' # 1

def dooropen(): # 2return door == 'Open'

3.9. Processes 271


class Doorman(Process): # 3""" Doorman opens the door"""

def openthedoor(self):""" He will open the door when he arrives"""global dooryield hold, self, expovariate(1.0 / 10.0) # 4door = 'Open'print("%7.4f Doorman: Ladies and "

"Gentlemen! You may all enter." % (now()))




class Customer(Process): # 5""" Customer arrives, is served and leaves """

def visit(self, timeInBank=10):arrive = now()if dooropen():

msg = ' and the door is open.'else:

msg = ' but the door is shut.'print("%7.4f %s: Here I am%s" % (now(), self.name, msg))

yield waituntil, self, dooropen # 6

print("%7.4f %s: I can go in!" % (now(), self.name))wait = now() - arriveprint("%7.4f %s: Waited %6.3f" % (now(), self.name, wait))

yield request, self, countertib = expovariate(1.0 / timeInBank)yield hold, self, tibyield release, self, counter


# Experiment data -------------------------

maxTime = 2000.0 # minutescounter = Resource(1, name="Clerk")

# Model ----------------------------------

def model(SEED=393939):seed(SEED)



initialize()door = 'Shut'doorman = Doorman() # 7activate(doorman, doorman.openthedoor()) # 8source = Source()activate(source,

source.generate(number=5, rate=0.1), at=0.0)simulate(until=400.0)

# Experiment ----------------------------------model()

The output from a run for this programs shows how the first customer has to wait until the door is opened.

0.0000 Customer00: Here I am but the door is shut.1.1489 Doorman: Ladies and Gentlemen! You may all enter.1.1489 Customer00: I can go in!1.1489 Customer00: Waited 1.1496.5691 Customer00: Finished8.3438 Customer01: Here I am and the door is open.8.3438 Customer01: I can go in!8.3438 Customer01: Waited 0.000

15.5704 Customer02: Here I am and the door is open.15.5704 Customer02: I can go in!15.5704 Customer02: Waited 0.00021.2664 Customer03: Here I am and the door is open.21.2664 Customer03: I can go in!21.2664 Customer03: Waited 0.00021.9473 Customer04: Here I am and the door is open.21.9473 Customer04: I can go in!21.9473 Customer04: Waited 0.00027.6401 Customer01: Finished56.5248 Customer02: Finished57.3640 Customer03: Finished77.3587 Customer04: Finished

3.9.3 Wait for the doorman to give a signal: waitevent

Customers arrive at random, some of them getting to the bank before the door is open. This is controlled by anautomatic machine called the doorman which opens the door only at intervals of 30 minutes (it is a very secure bank).The customers wait for the door to be opened and all those waiting enter and proceed to the counter. The door is closedbehind them.

This model uses the yield waitevent command which requires a SimEvent to be defined (label #1). TheDoorman class is defined at label #2 and the doorman is created and activated at labels #7 and #8. The doormanwaits for a fixed time (label #3) and then tells the customers that the door is open. This is achieved on label #4 bysignalling the dooropen event.

The Customer class is defined at label #5 and in its PEM, when a customer arrives, he prints out Here I am. If thedoor is still closed, he adds “but the door is shut‘ and settles down to wait for the door to be opened using the yieldwaitevent command (label #6). When the door is opened by the doorman (that is, he sends the dooropen.signal() the customer and any others waiting may proceed.

""" bank13: Wait for the doorman to give a signal: *waitevent*"""from SimPy.Simulation import *

3.9. Processes 273


from random import *


dooropen = SimEvent("Door Open") # 1

class Doorman(Process): # 2""" Doorman opens the door"""

def openthedoor(self):""" He will opens the door at fixed intervals"""for i in range(5):

yield hold, self, 30.0 # 3dooropen.signal() # 4print("%7.4f You may enter" % (now()))




class Customer(Process): # 5""" Customer arrives, is served and leaves """

def visit(self, timeInBank=10):arrive = now()

if dooropen.occurred:msg = '.'

else:msg = ' but the door is shut.'

print("%7.4f %s: Here I am%s" % (now(), self.name, msg))yield waitevent, self, dooropen

print("%7.4f %s: The door is open!" % (now(), self.name))

wait = now() - arriveprint("%7.4f %s: Waited %6.3f" % (now(), self.name, wait))

yield request, self, countertib = expovariate(1.0 / timeInBank)yield hold, self, tibyield release, self, counter


# Experiment data -------------------------

maxTime = 400.0 # minutes



counter = Resource(1, name="Clerk")

# Model ----------------------------------


initialize()doorman = Doorman() # 7activate(doorman, doorman.openthedoor()) # 8source = Source()activate(source,


# Experiment ----------------------------------

model()

An output run for this programs shows how the first three customers have to wait until the door is opened.

0.0000 Customer00: Here I am but the door is shut.13.6767 Customer01: Here I am but the door is shut.13.9068 Customer02: Here I am but the door is shut.30.0000 You may enter30.0000 Customer02: The door is open!30.0000 Customer02: Waited 16.09330.0000 Customer01: The door is open!30.0000 Customer01: Waited 16.32330.0000 Customer00: The door is open!30.0000 Customer00: Waited 30.00034.0411 Customer03: Here I am but the door is shut.40.8095 Customer04: Here I am but the door is shut.55.4721 Customer02: Finished57.2363 Customer01: Finished60.0000 You may enter60.0000 Customer04: The door is open!60.0000 Customer04: Waited 19.19060.0000 Customer03: The door is open!60.0000 Customer03: Waited 25.95977.0409 Customer00: Finished90.0000 You may enter104.8327 Customer04: Finished118.4142 Customer03: Finished120.0000 You may enter150.0000 You may enter

3.9. Processes 275


3.10 Monitors

Monitors (and Tallys) are used to track and record values in a simulation. They store a list of [time,value] pairs, onepair being added whenever the observe method is called. A particularly useful characteristic is that they continue toexist after the simulation has been completed. Thus further analysis of the results can be carried out.

Monitors have a set of simple statistical methods such as mean and var to calculate the average and variance of theobserved values – useful in estimating the mean delay, for example.

They also have the timeAverage method that calculates the time-weighted average of the recorded values. Itdetermines the total area under the time~value graph and divides by the total time. This is useful for estimating theaverage number of customers in the bank, for example. There is an important caveat in using this method. To estimatethe correct time average you must certainly observe the value (say the number of customers in the system) wheneverit changes (as well as at any other time you wish) but, and this is important, observing the new value. The old valuewas recorded earlier. In practice this means that if we wish to observe a changing value, n, using the Monitor, Mon,we must keep to the the following pattern:

n = n+1Mon.observe(n,now())

Thus you make the change (not only increases) and then observe the new value. Of course the simulation time now()has not changed between the two statements.

3.10.1 Plotting a Histogram of Monitor results

A Monitor can construct a histogram from its data using the histogram method. In this model we monitor the timein the system for the customers. This is calculated for each customer at label #2, using the arrival time saved at label#1. We create the Monitor, Mon, at label #4 and the times are observed at label #3.

The histogram is constructed from the Monitor, after the simulation has finished, at label #5. The SimPy SimPlotpackage allows simple plotting of results from simulations. Here we use the SimPlot plotHistogram method. Theplotting routines appear between labels #6 and #7. The plotHistogram call is in label #7.

"""bank17: Plotting a Histogram of Monitor results"""from SimPy.Simulation import *from SimPy.SimPlot import *from random import expovariate, seed






def visit(self, timeInBank):arrive = now() # 1



# print("%8.4f %s: Arrived " % (now(), self.name))

yield request, self, counter# print("%8.4f %s: Got counter " % (now(), self.name))tib = expovariate(1.0 / timeInBank)yield hold, self, tibyield release, self, counter

# print("%8.4f %s: Finished " % (now(), self.name))t = now() - arrive # 2Mon.observe(t) # 3

# Experiment data -------------------------

maxTime = 400.0 # minutescounter = Resource(1, name="Clerk")Mon = Monitor('Time in the Bank') # 4N = 0

# Model ----------------------------------


initialize()source = Source()activate(source,


# Experiment ----------------------------------model()Histo = Mon.histogram(low=0.0, high=200.0, nbins=20) # 5

plt = SimPlot() # 6plt.plotHistogram(Histo, xlab='Time (min)', # 7

title="Time in the Bank",color="red", width=2)

plt.mainloop() # 8

3.10.2 Plotting from Resource Monitors

Like all Monitors, waitMon and actMon in a monitored Resource contain information that enables us to graph theoutput. Alternative plotting packages can be used; here we use the simple SimPy.SimPlot package just to graphthe number of customers waiting for the counter. The program is a simple modification of the one that uses a monitoredResource.

The SimPlot package is imported at #3. No major changes are made to the main part of the program except thatI commented out the print statements. The changes occur in the model routine from #3 to #5.The simulation nowgenerates and processes 20 customers (#4). model does not return a value but the Monitors of the counter Resourcestill exist when the simulation has terminated.

The additional plotting actions take place from #6 to #9. Lines #7 and #8 construct a step plot and graphs the numberin the waiting queue as a function of time. waitMon is primarily a list of [time,value] pairs which the plotStep

3.10. Monitors 277


method of the SimPlot object, plt uses without change. On running the program the graph is plotted; the user has toterminate the plotting mainloop on the screen.

"""bank16: Plotting from Resource Monitors"""from SimPy.Simulation import *from SimPy.SimPlot import * # 1from random import expovariate, seed






def visit(self, timeInBank):arrive = now()# print("%8.4f %s: Arrived " % (now(), self.name))

yield request, self, counter# print("%8.4f %s: Got counter " % (now(), self.name))tib = expovariate(1.0 / timeInBank)yield hold, self, tibyield release, self, counter

# print("%8.4f %s: Finished " % (now(), self.name))

# Experiment data -------------------------

maxTime = 400.0 # minutescounter = Resource(1, name="Clerk", monitored=True)

# Model -----------------------------------

def model(SEED=393939): # 3seed(SEED)initialize()source = Source()activate(source, # 4

source.generate(number=20, rate=0.1), at=0.0)simulate(until=maxTime) # 5

# Experiment -----------------------------------

model()

plt = SimPlot() # 6



plt.plotStep(counter.waitMon, # 7color="red", width=2) # 8

plt.mainloop() # 9

3.11 Acknowledgements

I thank Klaus Muller, Bob Helmbold, Mukhlis Matti and the other developers and users of SimPy for improving thisdocument by sending their comments. I would be grateful for any further corrections or suggestions. Please send themto: vignaux at users.sourceforge.net.

3.12 References


• SimPy homepage: https://github.com/SimPyClassic/SimPyClassic

• The Bank:

3.13 The Bank (Object Oriented version)

3.13.1 Introduction

SimPy is used to develop a simple simulation of a bank with a number of tellers. This Python package providesProcesses to model active components such as messages, customers, trucks, and planes. It has three classes to modelfacilities where congestion might occur: Resources for ordinary queues, Levels for the supply of quantities of material,and Stores for collections of individual items. Only examples of Resources are described here. It also providesMonitors and Tallys to record data like queue lengths and delay times and to calculate simple averages. It uses thestandard Python random package to generate random numbers.

Starting with SimPy 2.0 an object-oriented programmer’s interface was added to the package and it is this version thatis described here. It is quite compatible with the procedural approach. The object-oriented interface, however, cansupport the process of developing and extending a simulation model better than the procedural approach.

SimPy can be obtained from: https://github.com/SimPyClassic/SimPyClassic. This tutorial is best read with the SimPyManual and CheatsheetOO at your side for reference.

Before attempting to use SimPy you should be familiar with the Python language. In particular you should be able touse classes. Python is free and available for most machine types. You can find out more about it at the Python website. SimPy is compatible with Python version 2.3 and later.

3.13.2 A single Customer

In this tutorial we model a simple bank with customers arriving at random. We develop the model step-by-step,starting out simply, and producing a running program at each stage. The programs we develop are available withoutline numbers and ready to go, in the bankprograms_OO directory. Please copy them, run them and improve them- and in the tradition of open-source software suggest your modifications to the SimPy users list. Object-orentedversions of all the models are included in the same directory.

A simulation should always be developed to answer a specific question; in these models we investigate how changingthe number of bank servers or tellers might affect the waiting time for customers.

3.11. Acknowledgements 279









A Customer arriving at a fixed time

We first model a single customer who arrives at the bank for a visit, looks around at the decor for a time and thenleaves. There is no queueing. First we will assume his arrival time and the time he spends in the bank are fixed.

Examine the following listing which is a complete runnable Python script, except for the line numbers. We usecomments to divide the script up into sections. This makes for clarity later when the programs get more complicated.Line 1 is a normal Python documentation string; line 2 makes available the SimPy constructs needed for this model:the Simulation class, the Process class, and the hold verb.

We define a Customer class derived from the SimPy Process class. We create a Customer object, c who arrivesat the bank at simulation time 5.0 and leaves after a fixed time of 10.0 minutes. The Customer class definition,lines 5-12, defines our customer class and has the required generator method (called visit) (line 9) having a yieldstatement (line 11)). Such a method is called a Process Execution Method (PEM) in SimPy.

The customer’s visit PEM, lines 9-12, models his activities. When he arrives (it will turn out to be a ‘he’ in thismodel), he will print out the simulation time, self.sim.now(), and his name (line 10). self.sim is a referenceto the BankModel simulation object where this customer exists. Every Process instance is linked to the simulationin which it is created by assigning to its sim parameter when it is created (see line 19). The method now() can beused at any time in the simulation to find the current simulation time though it cannot be changed by the programmer.The customer’s name will be set when the customer is created later in the script (line 19).

He then stays in the bank for a fixed simulation time timeInBank (line 11). This is achieved by the yield hold,self,timeInBank statement. This is the first of the special simulation commands that SimPy offers.

After a simulation time of timeInBank, the program’s execution returns to the line after the yield statement, line12. The customer then prints out the current simulation time and his name. This completes the declaration of theCustomer class.

Though we do not do it here, it is also possible to define an __init__() method for a Process if you needto give the customer any attributes. Bear in mind that such an __init__ method must first call Process.__init__(self) and can then initialize any instance variables needed.

Lines 6 to 21 define a class BankModel, composing a model of a bank from the Simulation class, a Customerclass and the global experiment data. The definition class BankModel(Simulation) gives an instance of aBankModel all the attributes of class Simulation. (In OO terms, BankModel inherits from Simulation.)Any instance of BankModel is a Simulation instance. This gives a BankModel its own event list and thus itsown time axis. Also, it allows a BankModel instance to activate processes and to start the execution of a simulationon its time axis.

Lines 17 to 21 define a run method; when called, it results in the execution of a BankModel instance, i.e. theperformance of a simulation experiment. Line 18 initializes this simulation, i.e. it creates a new event list. L.19creates a Customer object. The parameter assignment sim = self ties the customer instance to this and only thissimulation. The customer does not exist outside this simulation. L.20 activates the customer’s visit process (PEM).Finally the call of simulate(until=maxTime) in line 24 starts the simulation. It will run until the simulationtime is maxTime unless stopped beforehand either by the stopSimulation() command or by running out ofevents to execute (as will happen here). maxTime was set to 100.0 in line 25.

Note: If model classes like the‘‘BankModel‘‘ are to be given any other attributes, they must have an __init__method in which these attributes are assigned with the syntax self.attrib1 = . . .. Such an __init__method must first call Simulation.__init__(self) to also initialize the Simulation class from which themodel inherits.

The simulation model is executed by line 32. BankModel() constructs the model, and .run() executes it. This isjust a short form of:



bM = BankModel()bM.run()

""" bank01_OO: The single non-random Customer """from SimPy.Simulation import Simulation, Process, hold # 1# Model components -----------------------------


def visit(self, timeInBank):print("%2.1f %s Here I am" %

(self.sim.now(), self.name))yield hold, self, timeInBankprint("%2.1f %s I must leave" %

(self.sim.now(), self.name))

# Model ----------------------------------------

class BankModel(Simulation): # 3def run(self):

""" An independent simulation object """c = Customer(name="Klaus", sim=self) # 4self.activate(c, c.visit(timeInBank), at=tArrival)self.simulate(until=maxTime) # 5

# Experiment data ------------------------------maxTime = 100.0 # minutes #6timeInBank = 10.0 # minutestArrival = 5.0 # minutes

# Experiment -----------------------------------mymodel = BankModel() # 7mymodel.run() # 8



A Customer arriving at random

Now we extend the model to allow our customer to arrive at a random simulated time though we will keep the time inthe bank at 10.0, as before.

The change occurs in line 3 of the program and in lines 19, 21, 23, 31 and 35. In line 3 we import from the standardPython random module to give us expovariate to generate the random time of arrival. We also import the seedfunction to initialize the random number stream to allow control of the random numbers. The run method is givena parameter aseed for the initial seed (line 19) .In line 31 we provide an initial seed of 99999. An exponentialrandom variate, t, is generated in line 23. Note that the Python Random module’s expovariate function uses the

3.13. The Bank (Object Oriented version) 281


rate (here, 1.0/tMeanArrival) as the argument. The generated random variate, t, is used in line 24 as the atargument to the activate call. tMeanArrival is assigned a value of 5.0 minutes at line 31.

In line 35, the BankModel entity is generated and its run function called with parameter assignmentaseed=seedVal.

""" bank05_OO: The single Random Customer """from SimPy.Simulation import Simulation, Process, holdfrom random import expovariate, seed


class Customer(Process):""" Customer arrives at a random time,

looks around and then leaves """

def visit(self, timeInBank):print("%f %s Here I am" % (self.sim.now(), self.name))yield hold, self, timeInBankprint("%f %s I must leave" % (self.sim.now(), self.name))

# Model -----------------------------------

class BankModel(Simulation):def run(self, aseed):

""" PEM """seed(aseed)c = Customer(name="Klaus", sim=self)t = expovariate(1.0 / tMeanArrival)self.activate(c, c.visit(timeInBank), at=t)self.simulate(until=maxTime)

# Experiment data -------------------------

maxTime = 100.0 # minutestimeInBank = 10.0 # minutestMeanArrival = 5.0 # minutesseedVal = 99999

# Experiment ------------------------------

mymodel = BankModel()mymodel.run(aseed=seedVal)

The result is shown below. The customer now arrives at time 10.5809. Changing the seed value would change thattime.


The display looks pretty untidy. In the next example I will try and make it tidier.



3.13.3 More Customers

Our simulation does little so far. To consider a simulation with several customers we return to the simple deterministicmodel and add more Customers.

The program is almost as easy as the first example (A Customer arriving at a fixed time). The main change is in lines19-24 where we create, name, and activate three customers. We also increase the maximum simulation time to 400(line 29 and referred to in line 25). Observe that we need only one definition of the Customer class and create severalobjects of that class. These will act quite independently in this model.

Each customer stays for a different timeInBank so, instead of setting a common value for this we set it for eachcustomer. The customers are started at different times (using at=). Tony's activation time occurs before Klaus's,so Tony will arrive first even though his activation statement appears later in the script.

As promised, the print statements have been changed to use Python string formatting (lines 10 and 12). The statementslook complicated but the output is much nicer.

""" bank02_OO: More Customers """from SimPy.Simulation import Simulation, Process, hold

# Model components ------------------------class Customer(Process):

""" Customer arrives, looks around and leaves """

def visit(self, timeInBank=0):print("%7.4f %s: Here I am" % (self.sim.now(), self.name))yield hold, self, timeInBankprint("%7.4f %s: I must leave" % (self.sim.now(), self.name))

# Model -----------------------------------class BankModel(Simulation):

def run(self):""" PEM """c1 = Customer(name="Klaus", sim=self)self.activate(c1, c1.visit(timeInBank=10.0), at=5.0)c2 = Customer(name="Tony", sim=self)self.activate(c2, c2.visit(timeInBank=7.0), at=2.0)c3 = Customer(name="Evelyn", sim=self)self.activate(c3, c3.visit(timeInBank=20.0), at=12.0)self.simulate(until=maxTime)

# Experiment data -------------------------


# Experiment ------------------------------

mymodel = BankModel()mymodel.run()

The trace produced by the program is shown below. Again the simulation finishes before the 400.0 specified in thesimulate call.





Many Customers

Another change will allow us to have more customers. As it is tedious to give a specially chosen name to each one, wewill call them Customer00, Customer01,... and use a separate Source class to create and activate them.To make things clearer we do not use the random numbers in this model.

The following listing shows the new program. Lines 6-13 define a Source class. Its PEM, here called generate,is defined in lines 9-13. This PEM has a couple of arguments: the number of customers to be generated and the TimeBetween Arrivals, TBA. It consists of a loop that creates a stream of numbered Customers from 0 to (number-1),inclusive. We create a customer and give it a name in line 11. The parameter assignment sim = self.sim ties thecustomers to the BankModel to which the Source belongs. The customer is then activated at the current simulationtime (the final argument of the activate statement is missing so that the default value of self.sim.now(), thecurrent simulation time for the instance of BankModel, is used as the time; here, it is 0.0). We also specify howlong the customer is to stay in the bank. To keep it simple, all customers stay exactly 12 minutes. After each newcustomer is activated, the Source holds for a fixed time (yield hold,self,TBA) before creating the next one(line 13).

class BankModel(Simulation) (line 24) provides a run method which executes this model consisting of acustomer source and the global data. As BankModel inherits from Simulation, it has its own event list whichgets initialized as empty in line 26.

A Source, s, is created in line 27 and activated at line 28 where the number of customers to be generated is setto maxNumber = 5 and the interval between customers to ARRint = 10.0. The parameter assignment sim =self links the Source process to this BankModel instance. Once started at time 0.0, s creates customers atintervals and each customer then operates independently of the others.

In line 40, a BankModel object is created and its run method executed:

""" bank03_OO: Many non-random Customers """from SimPy.Simulation import Simulation, Process, hold

# Model components ------------------------class Source(Process):

""" Source generates customers regularly """

def generate(self, number, TBA):for i in range(number):

c = Customer(name="Customer%02d" % (i), sim=self.sim)self.sim.activate(c, c.visit(timeInBank=12.0))yield hold, self, TBA


def visit(self, timeInBank):print("%7.4f %s: Here I am" % (self.sim.now(), self.name))yield hold, self, timeInBankprint("%7.4f %s: I must leave" % (self.sim.now(), self.name))




def run(self):""" PEM """s = Source(sim=self)self.activate(s, s.generate(number=maxNumber,

TBA=ARRint), at=0.0)self.simulate(until=maxTime)

# Experiment data -------------------------

maxNumber = 5maxTime = 400.0 # minutesARRint = 10.0 # time between arrivals, minutes

# Experiment ------------------------------

mymodel = BankModel()mymodel.run()

The output is:

0.0000 Customer00: Here I am10.0000 Customer01: Here I am12.0000 Customer00: I must leave20.0000 Customer02: Here I am22.0000 Customer01: I must leave30.0000 Customer03: Here I am32.0000 Customer02: I must leave40.0000 Customer04: Here I am42.0000 Customer03: I must leave52.0000 Customer04: I must leave

Many Random Customers

We now extend this model to allow arrivals at random. In simulation this is usually interpreted as meaning that thetimes between customer arrivals are distributed as exponential random variates. There is little change in our program,we use a Source object, as before.

The exponential random variate is generated in line 14 with meanTBA as the mean Time Between Arrivals and usedin line 15. Note that this parameter is not exactly intuitive. As already mentioned, the Python expovariate methoduses the rate of arrivals as the parameter not the average interval between them. The exponential delay between twoarrivals gives pseudo-random arrivals. In this model the first customer arrives at time 0.0.

The seed method is called to initialize the random number stream in the run routine of BankModel (line 30).It uses the value provided by parameter aseed. It is possible to leave this call out but if we wish to do seriouscomparisons of systems, we must have control over the random variates and therefore control over the seeds. Then wecan run identical models with different seeds or different models with identical seeds. We provide the seeds as controlparameters of the run. Here a seed is assigned in line 41 but it is clear it could have been read in or manually enteredon an input form.

The BankModel is generated in line 45 and its run method called with the seed value as parameter.

""" bank06: Many Random Customers """from SimPy.Simulation import Simulation, Process, holdfrom random import expovariate, seed



# Model components ------------------------class Source(Process):

""" Source generates customers at random """


c = Customer(name="Customer%02d" % (i), sim=self.sim)self.sim.activate(c, c.visit(timeInBank=12.0))t = expovariate(1.0 / meanTBA)yield hold, self, t

class Customer(Process):""" Customer arrives, looks round and leaves """

def visit(self, timeInBank=0):print("%7.4f %s: Here I am" % (self.sim.now(), self.name))yield hold, self, timeInBankprint("%7.4f %s: I must leave" % (self.sim.now(), self.name))


def run(self, aseed):""" PEM """seed(aseed)s = Source(name='Source', sim=self)self.activate(s, s.generate(number=maxNumber,

meanTBA=ARRint), at=0.0)self.simulate(until=maxTime)

# Experiment data -------------------------

maxNumber = 5maxTime = 400.0 # minutesARRint = 10.0 # mean arrival interval, minutesseedVal = 99999

# Experiment ------------------------------


This generates the following output:

0.0000 Customer00: Here I am1.2839 Customer01: Here I am4.9842 Customer02: Here I am

12.0000 Customer00: I must leave13.2839 Customer01: I must leave16.9842 Customer02: I must leave35.5432 Customer03: Here I am47.5432 Customer03: I must leave48.9918 Customer04: Here I am60.9918 Customer04: I must leave



3.13.4 A Service counter

So far, the model bank has been more like an art gallery, the customers entering, looking around, and leaving. Nowthey are going to require service from the bank clerk. We extend the model to include a service counter which willbe modelled as an object of SimPy’s Resource class with a single resource unit. The actions of a Resource aresimple: a customer requests a unit of the resource (a clerk). If one is free he gets service (and removes the unit,i.e., makes it busy). If there is no free clerk the customer joins the queue (managed by the resource object) until itis his turn to be served. As each customer completes service and releases the unit, the clerk automatically startsserving the next in line. This is done by reactivating that customer’s process where it had been blocked.

One Service counter

As this model is built with the Resource class from SimPy.Simulation, it and the related request andrelease verbs are mported, in addition to the imports made in the previous programs (line 2).

The service counter is created as a Resource attribute self.k of the BankModel (line 39). The resource exists inthe BankModel, and this is indicated by the parameter assignment sim = self. The Source PEM generatecan access this attribute by self.sim.k, its BankModel’s resource attribute (line 14).

The actions involving the Counter referred to by the parameter res in the customer’s PEM are:

• the yield request statement in line 25. If the server is free then the customer can start service immediatelyand the code moves on to line 26. If the server is busy, the customer is automatically queued by the Resource.When it eventually comes available the PEM moves on to line 26.

• the yield hold statement in line 28 where the operation of the service counter is modelled. Here the servicetime is a fixed timeInBank. During this period the customer is being served and the resource (the counter) isbusy.

• the yield release statement in line 29. The current customer completes service and the service counterbecomes available for any remaining customers in the queue.

Observe that the service counter is used with the pattern (yield request..; yield hold..; yieldrelease..).

To show the effect of the service counter on the activities of the customers, I have added line 22 to record when thecustomer arrived and line 26 to record the time between arrival in the bank and starting service. Line 26 is after theyield request command and will be reached only when the request is satisfied. It is before the yield holdthat corresponds to the start of service. The variable wait will record how long the customer waited and will be 0 ifhe received service at once. This technique of saving the arrival time in a variable is common. So the print statementalso prints out how long the customer waited in the bank before starting service.

""" bank07_OO: One Counter,random arrivals """from SimPy.Simulation import (Simulation, Process, hold, Resource, request,

release)from random import expovariate, seed




c = Customer(name="Customer%02d" % (i), sim=self.sim)self.sim.activate(c, c.visit(timeInBank=12.0,

res=self.sim.k))



t = expovariate(1.0 / meanTBA)yield hold, self, t


def visit(self, timeInBank=0, res=None):arrive = self.sim.now() # arrival timeprint("%8.3f %s: Here I am" % (self.sim.now(), self.name))

yield request, self, reswait = self.sim.now() - arrive # waiting timeprint("%8.3f %s: Waited %6.3f" % (self.sim.now(), self.name, wait))yield hold, self, timeInBankyield release, self, res

print("%8.3f %s: Finished" % (self.sim.now(), self.name))

# Model -----------------------------------


""" PEM """seed(aseed)self.k = Resource(name="Counter", unitName="Clerk", sim=self)s = Source('Source', sim=self)self.activate(s, s.generate(number=maxNumber, meanTBA=ARRint), at=0.0)self.simulate(until=maxTime)

# Experiment data -------------------------

maxNumber = 5maxTime = 400.0 # minutesARRint = 10.0 # mean, minutesseedVal = 99999

# Experiment ------------------------------


Examining the trace we see that the first two customers get instant service but the others have to wait. We still onlyhave five customers (line 44) so we cannot draw general conclusions.

0.000 Customer00: Here I am0.000 Customer00: Waited 0.0001.284 Customer01: Here I am4.984 Customer02: Here I am12.000 Customer00: Finished12.000 Customer01: Waited 10.71624.000 Customer01: Finished24.000 Customer02: Waited 19.01635.543 Customer03: Here I am36.000 Customer02: Finished36.000 Customer03: Waited 0.457



48.000 Customer03: Finished48.992 Customer04: Here I am48.992 Customer04: Waited 0.00060.992 Customer04: Finished

A server with a random service time

This is a simple change to the model in that we retain the single service counter but make the customer service timea random variable. As is traditional in the study of simple queues we first assume an exponential service time and setthe mean to timeInBank.

The service time random variable, tib, is generated in line 26 and used in line 27. The argument to be used in thecall of expovariate is not the mean of the distribution, timeInBank, but is the rate 1.0/timeInBank.

We have put together the exeriment data by defining a number of appropriate variables and giving them values. Theseare in lines 44 to 48.

""" bank08_OO: A counter with a random service time """from SimPy.Simulation import (Simulation, Process, Resource, hold, request,





c = Customer(name="Customer%02d" % (i), sim=self.sim)self.sim.activate(c, c.visit(b=self.sim.k))t = expovariate(1.0 / meanTBA)yield hold, self, t


def visit(self, b):arrive = self.sim.now()print("%8.4f %s: Here I am " % (self.sim.now(), self.name))yield request, self, bwait = self.sim.now() - arriveprint("%8.4f %s: Waited %6.3f" % (self.sim.now(), self.name, wait))tib = expovariate(1.0 / timeInBank)yield hold, self, tibyield release, self, bprint("%8.4f %s: Finished " % (self.sim.now(), self.name))

# Model -----------------------------------


""" PEM """seed(aseed)



self.k = Resource(name="Counter", unitName="Clerk", sim=self)s = Source('Source', sim=self)self.activate(s, s.generate(number=maxNumber, meanTBA=ARRint), at=0.0)self.simulate(until=maxTime)

# Experiment data -------------------------

maxNumber = 5maxTime = 400.0 # minutestimeInBank = 12.0 # mean, minutesARRint = 10.0 # mean, minutesseedVal = 99999

# Experiment ------------------------------


And the output:

0.0000 Customer00: Here I am0.0000 Customer00: Waited 0.0001.2839 Customer01: Here I am4.4403 Customer00: Finished4.4403 Customer01: Waited 3.156

20.5786 Customer01: Finished31.8430 Customer02: Here I am31.8430 Customer02: Waited 0.00034.5594 Customer02: Finished36.2308 Customer03: Here I am36.2308 Customer03: Waited 0.00041.4313 Customer04: Here I am67.1315 Customer03: Finished67.1315 Customer04: Waited 25.70087.9241 Customer04: Finished

This model with random arrivals and exponential service times is an example of an M/M/1 queue and could rathereasily be solved analytically to calculate the steady-state mean waiting time and other operating characteristics. (Butnot so easily solved for its transient behavior.)

3.13.5 Several Service Counters

When we introduce several counters we must decide on a queue discipline. Are customers going to make one queue orare they going to form separate queues in front of each counter? Then there are complications - will they be allowedto switch lines (jockey)? We first consider a single queue with several counters and later consider separate isolatedqueues. We will not look at jockeying.

Several Counters but a Single Queue

Here we model a bank whose customers arrive randomly and are to be served at a group of counters, taking a randomtime for service, where we assume that waiting customers form a single first-in first-out queue.

The only difference between this model and the single-server model is in line 37. We have provided two countersby increasing the capacity of the counter resource to 2. This value is set in line 50 (Nc = 2). These units of the



resource correspond to the two counters. Because both clerks cannot be called Karen, we have used a general nameof Clerk as resource unit.

""" bank09_OO: Several Counters but a Single Queue """from SimPy.Simulation import (Simulation, Process, Resource, hold, request,





c = Customer(name="Customer%02d" % (i), sim=self.sim)self.sim.activate(c, c.visit(b=self.sim.k))t = expovariate(1.0 / meanTBA)yield hold, self, t


def visit(self, b):arrive = self.sim.now()print("%8.4f %s: Here I am " % (self.sim.now(), self.name))yield request, self, bwait = self.sim.now() - arriveprint("%8.4f %s: Waited %6.3f" % (self.sim.now(), self.name, wait))tib = expovariate(1.0 / timeInBank)yield hold, self, tibyield release, self, bprint("%8.4f %s: Finished " % (self.sim.now(), self.name))

# Model -----------------------------------


""" PEM """seed(aseed)self.k = Resource(capacity=Nc, name="Counter", unitName="Clerk",

sim=self)s = Source('Source', sim=self)self.activate(s, s.generate(number=maxNumber, meanTBA=ARRint), at=0.0)self.simulate(until=maxTime)

# Experiment data -------------------------

maxNumber = 5 # of customersmaxTime = 400.0 # minutestimeInBank = 12.0 # mean, minutesARRint = 10.0 # mean, minutesNc = 2 # of clerks/countersseedVal = 99999



# Experiment ------------------------------mymodel = BankModel()mymodel.run(aseed=seedVal)

The waiting times in this model are much shorter than those for the single service counter. For example, the waitingtime for Customer02 has been reduced from 51.213 to 12.581 minutes. Again we have too few customersprocessed to draw general conclusions.

0.0000 Customer00: Here I am0.0000 Customer00: Waited 0.0001.2839 Customer01: Here I am1.2839 Customer01: Waited 0.0004.4403 Customer00: Finished

17.4222 Customer01: Finished31.8430 Customer02: Here I am31.8430 Customer02: Waited 0.00034.5594 Customer02: Finished36.2308 Customer03: Here I am36.2308 Customer03: Waited 0.00041.4313 Customer04: Here I am41.4313 Customer04: Waited 0.00062.2239 Customer04: Finished67.1315 Customer03: Finished

Several Counters with individual queues

Each counter is now assumed to have its own queue. The programming is more complicated because the customer hasto decide which queue to join. The obvious technique is to make each counter a separate resource and it is useful tomake a list of resource objects (line 56).

In practice, a customer will join the shortest queue. So we define the Python function, NoInSystem(R) (lines 17-19) which returns the sum of the number waiting and the number being served for a particular counter, R. This functionis used in line 28 to list the numbers at each counter. It is then easy to find which counter the arriving customer shouldjoin. We have also modified the trace printout, line 29 to display the state of the system when the customer arrives.We choose the shortest queue in lines 30-33 (the variable choice).

The rest of the program is the same as before.

""" bank10_OO: Several Counters with individual queues"""from SimPy.Simulation import (Simulation, Process, Resource, hold, request,





c = Customer(name="Customer%02d" % (i,), sim=self.sim)self.sim.activate(c, c.visit(counters=self.sim.kk))t = expovariate(1.0 / interval)yield hold, self, t



def NoInSystem(R):""" Total number of customers in the resource R"""return (len(R.waitQ) + len(R.activeQ))

class Customer(Process):""" Customer arrives, chooses the shortest queue

is served and leaves"""

def visit(self, counters):arrive = self.sim.now()Qlength = [NoInSystem(counters[i]) for i in range(Nc)]print("%7.4f %s: Here I am. %s" % (self.sim.now(), self.name, Qlength))for i in range(Nc):

if Qlength[i] == 0 or Qlength[i] == min(Qlength):choice = i # the chosen queue numberbreak

yield request, self, counters[choice]wait = self.sim.now() - arriveprint("%7.4f %s: Waited %6.3f" % (self.sim.now(), self.name, wait))tib = expovariate(1.0 / timeInBank)yield hold, self, tibyield release, self, counters[choice]

print("%7.4f %s: Finished" % (self.sim.now(), self.name))

# Model -----------------------------------


""" PEM """seed(aseed)self.kk = [Resource(name="Clerk0", sim=self),

Resource(name="Clerk1", sim=self)]s = Source('Source', sim=self)self.activate(s, s.generate(number=maxNumber, interval=ARRint), at=0.0)self.simulate(until=maxTime)

# Experiment data -------------------------

maxNumber = 5maxTime = 400.0 # minutestimeInBank = 12.0 # mean, minutesARRint = 10.0 # mean, minutesNc = 2 # number of countersseedVal = 787878

# Experiment ------------------------------


The results show how the customers choose the counter with the smallest number. Unlucky Customer02 who joins



the wrong queue has to wait until Customer00 finishes at time 55.067. There are, however, too few arrivals inthese runs, limited as they are to five customers, to draw any general conclusions about the relative efficiencies of thetwo systems.

0.0000 Customer00: Here I am. [0, 0]0.0000 Customer00: Waited 0.0009.7519 Customer00: Finished

12.0829 Customer01: Here I am. [0, 0]12.0829 Customer01: Waited 0.00025.9167 Customer02: Here I am. [1, 0]25.9167 Customer02: Waited 0.00038.2349 Customer03: Here I am. [1, 1]40.4032 Customer04: Here I am. [2, 1]43.0677 Customer02: Finished43.0677 Customer04: Waited 2.66444.0242 Customer01: Finished44.0242 Customer03: Waited 5.78960.1271 Customer03: Finished70.2500 Customer04: Finished

3.13.6 Monitors and Gathering Statistics

The traces of output that have been displayed so far are valuable for checking that the simulation is operating correctlybut would become too much if we simulate a whole day. We do need to get results from our simulation to answer theoriginal questions. What, then, is the best way to summarize the results?

One way is to analyze the traces elsewhere, piping the trace output, or a modified version of it, into a real statisticalprogram such as R for statistical analysis, or into a file for later examination by a spreadsheet. We do not have spaceto examine this thoroughly here. Another way of presenting the results is to provide graphical output.

SimPy offers an easy way to gather a few simple statistics such as averages: the Monitor and Tally classes. TheMonitor records the values of chosen variables as time series. (but see the comments in Final Remarks).

The Bank with a Monitor

We now demonstrate a Monitor that records the average waiting times for our customers. We return to the systemwith random arrivals, random service times and a single queue and remove the old trace statements. In practice, wewould make the printouts controlled by a variable, say, TRACE which is set in the experimental data (or read in as aprogram option - but that is a different story). This would aid in debugging and would not complicate the data analysis.We will run the simulations for many more arrivals.

In addition to the imports in the programs shown before, we now have to import the Monitor class (line 2).

A Monitor, wM, is created in line 37. We make the monitor an attribute of the BankModel by the assignment to self.wM. The monitor observes the waiting time mentioned in line 25. As the monitor is an attribute of the BankModelto which the customer belongs, self.sim.wM can refere to it. We run maxNumber = 50 customers (in the callof generate in line 39) and have increased maxTime to 1000.0 minutes.

""" bank11: The bank with a Monitor"""from SimPy.Simulation import Simulation, Process, Resource, Monitor, hold,\

request, releasefrom random import expovariate, seed






c = Customer(name="Customer%02d" % (i), sim=self.sim)self.sim.activate(c, c.visit(b=self.sim.k))t = expovariate(1.0 / interval)yield hold, self, t


def visit(self, b):arrive = self.sim.now()yield request, self, bwait = self.sim.now() - arriveself.sim.wM.observe(wait)tib = expovariate(1.0 / timeInBank)yield hold, self, tibyield release, self, b

# Model -----------------------------------


""" PEM """seed(aseed)self.k = Resource(capacity=Nc, name="Clerk", sim=self)self.wM = Monitor(sim=self)s = Source('Source', sim=self)self.activate(s, s.generate(number=maxNumber, interval=ARRint), at=0.0)self.simulate(until=maxTime)

# Experiment data -------------------------

maxNumber = 50maxTime = 1000.0 # minutestimeInBank = 12.0 # mean, minutesARRint = 10.0 # mean, minutesNc = 2 # number of countersseedVal = 99999

# Experiment -----------------------------

experi = BankModel()experi.run(aseed=seedVal)

# Result ----------------------------------

result = experi.wM.count(), experi.wM.mean()print("Average wait for %3d completions was %5.3f minutes." % result)

In previous programs, we have generated the BankModel anonymously. Here, we do it differently: we assign theBankModel object to the variable experi (line 53). This way, we can reference its monitor attribute by experi.



wM (line 58). The average waiting time for 50 customers in this 2-counter system is more reliable (i.e., less subjectto random simulation effects) than the times we measured before but it is still not sufficiently reliable for real-worlddecisions. We should also replicate the runs using different random number seeds. The result of this run is:


Multiple runs

This example demonstrates the power of the object-oriented approach. To get a number of independent measurementswe must replicate the runs using different random number seeds. Each replication must be independent of previousones both in the random numbers and in the data collection so capability of creating independent simulation modelswithin one program is useful.

We do a standard from SimPy.Simulation import ... at #1. We define Source and Customer as sub-classes of `Process. These differ in detail from the way they are defined in the classic procedure-oriented versionof SimPy. Each must refer to the simulation environment it is running in, here the sim argument. Thus the currenttime is returned by the self.sim.now() method (at #5

We define a BankModel as a sub-class of Simulation (at #8) and create an object, mymodel, of that class (at#14). A BankModel object has a run method and this is used for each independent replication (at #16). Note thatthe BankModel is only generated once (line 54). This is sufficient, as the run method freshly generates an emptyevent list, a new counter resource, a new monitor, and a new source. This way, all iterations are independent of eachother.

The random number seeds are stored in a list, seedVals and the for loop walks through this list and executesmymodel’s run method for each entry to get a set of replications.

""" bank12_OO: Multiple runs of the bank with a Monitor"""from SimPy.Simulation import Simulation, Process, \

Resource, Monitor, hold, request, release # 1from random import expovariate, seed



def generate(self, number, interval): # 2for i in range(number):

c = Customer(name="Customer%02d" % (i),sim=self.sim) # 3

self.sim.activate(c, c.visit(b=self.sim.k)) # 4t = expovariate(1.0 / interval)yield hold, self, t


def visit(self, b):arrive = self.sim.now() # 5yield request, self, bwait = self.sim.now() - arrive # 6self.sim.wM.observe(wait)tib = expovariate(1.0 / timeInBank) # 7yield hold, self, tib



yield release, self, b

# Simulation Model ----------------------------------

class BankModel(Simulation): # 8def run(self, aseed):

self.initialize() # 9seed(aseed)self.k = Resource(capacity=Nc, name="Clerk",

sim=self) # 10self.wM = Monitor(sim=self) # 11s = Source('Source', sim=self) # 12self.activate(s, s.generate(number=maxNumber,

interval=ARRint), at=0.0)self.simulate(until=maxTime) # 13return (self.wM.count(), self.wM.mean())

# Experiment data -------------------------maxNumber = 50maxTime = 2000.0 # minutestimeInBank = 12.0 # mean, minutesARRint = 10.0 # mean, minutesNc = 2 # number of countersseedVals = [393939, 31555999, 777999555, 319999771]

# Experiment/Result ----------------------------------

mymodel = BankModel() # 14for Sd in seedVals: # 15

mymodel.run(aseed=Sd) # 16moni = mymodel.wM # 17print("Avge wait for %3d completions was %6.2f min." %

(moni.count(), moni.mean()))

The results show some variation. Remember, though, that the system is still only operating for 50 customers so thesystem may not be in steady-state.


3.13.7 Final Remarks

This introduction is too long and the examples are getting longer. There is much more to say about simulation withSimPy but no space. I finish with a list of topics for further study:

• GUI input. Graphical input of simulation parameters could be an advantage in some cases. SimPy allows thisand programs using these facilities have been developed (see, for example, program MM1.py in the examplesin the SimPy distribution)

• Graphical Output. Similarly, graphical output of results can also be of value, not least in debugging simulationprograms and checking for steady-state conditions. SimPlot is useful here.



• Statistical Output. The Monitor class is useful in presenting results but more powerful methods of analysisare often needed. One solution is to output a trace and read that into a large-scale statistical system such as R.

• Priorities and Reneging in queues. SimPy allows processes to request units of resources under a priority queuediscipline (preemptive or not). It also allows processes to renege from a queue.

• Other forms of Resource Facilities. SimPy has two other resource structures: Levels to hold bulk commodi-ties, and Stores to contain an inventory of different object types.

• Advanced synchronization/scheduling commands. SimPy allows process synchronization by events and sig-nals.


I thank Klaus Muller, Bob Helmbold, Mukhlis Matti and other developers and users of SimPy for improving thisdocument by sending their comments. I would be grateful for further suggestions or corrections. Please send them to:vignaux at users.sourceforge.net.

3.13.9 References


• SimPy website: https://github.com/SimPyClassic/SimPyClassic

3.14 The Bank Tutorial (OO API) Part 2: More examples of SimPyClassic Simulation

Authors G A Vignaux, K G Muller

Date 2010 April

Release 2.3.3






3.14.1 OO API Bank Tutorial version

This manual is a rework of the Bank Tutorial Part 2. Its goal is to show how the simple tutorial models can be writtenin the advanced OO API.

Note: To contrast the OO API with the procedural SimPy API, the reader should read both “Bank Tutorial Part 2”documents side by side.

3.14.2 Introduction

The first Bank tutorial, The Bank, developed and explained a series of simulation models of a simple bank usingSimPy. In various models, customers arrived randomly, queued up to be served at one or several counters, modelledusing the Resource class, and, in one case, could choose the shortest among several queues. It demonstrated the use ofthe Monitor class to record delays and showed how a model() mainline for the simulation was convenient to executereplications of simulation runs.

In this extension to The Bank, I provide more examples of SimPy facilities for which there was no room and forsome that were developed since it was written. These facilities are generally more complicated than those introducedbefore. They include queueing with priority, possibly with preemption, reneging, plotting, interrupting, waiting untila condition occurs (waituntil) and waiting for events to occur.

Starting with SimPy 2.0 an object-oriented programmer’s interface was added to the package and it is this version thatis described here. It is quite compatible with the procedural approach. The object-oriented interface, however, cansupport the process of developing and extending a simulation model better than the procedural approach.

The programs are available without line numbers and ready to go, in directory bankprograms. Some have tracestatements for demonstration purposes, others produce graphical output to the screen. Let me encourage you to runthem and modify them for yourself.

SimPy itself can be obtained from: https://github.com/SimPyClassic/SimPyClassic. It is compatible with Pythonversion 2.7 onwards. The examples in this documentation run with SimPy version 1.5 and later.

This tutorial should be read with the SimPy Manual and CheatsheetOO at your side for reference.

3.14. The Bank Tutorial (OO API) Part 2: More examples of SimPy Classic Simulation 299




3.14.3 Priority Customers

In many situations there is a system of priority service. Those customers with high priority are served first, those withlow priority must wait. In some cases, preemptive priority will even allow a high-priority customer to interrupt theservice of one with a lower priority.

SimPy implements priority requests with an extra numerical priority argument in the yield request command,higher values meaning higher priority. For this to operate, the requested Resource must have been defined withqType=PriorityQ. This require importing the PriorityQ class from SimPy.Simulation.

Priority Customers without preemption

In the first example, we modify the program with random arrivals, one counter, and a fixed service time (like bank07.py in The Bank tutorial) to process a high priority customer. Warning: the seedVal value has been changed to98989 to make the story more exciting.

The modifications are to the definition of the counter where we change the qType and to the yield requestcommand in the visit PEM of the customer. We also need to provide each customer with a priority. Since thedefault is priority=0 this is easy for most of them.

To observe the priority in action, while all other customers have the default priority of 0, in lines 43 to 44 we createand activate one special customer, Guido, with priority 100 who arrives at time 23.0 (line 44). This is to ensure thathe arrives after Customer03.

The visit customer method has a new parameter, P=0 (line 20) which allows us to set the customer priority.

In lines 39 to 40 the BankModel ‘s resource attribute k named Counter is defined with qType=PriorityQ sothat we can request it with priority (line 25) using the statement yield request,self,self.sim.k,P

In line 23 we print out the number of customers waiting when each customer arrives.

""" bank20_OO: One counter with a priority customer """from SimPy.Simulation import (Simulation, Process, Resource, PriorityQ, hold,

request, release)from random import expovariate, seed




c = Customer(name="Customer%02d" % (i), sim=self.sim)self.sim.activate(c, c.visit(timeInBank=12.0, P=0))t = expovariate(1.0 / interval)yield hold, self, t


def visit(self, timeInBank=0, P=0):arrive = self.sim.now() # arrival timeNwaiting = len(self.sim.k.waitQ)print("%8.3f %s: Queue is %d on arrival" %

(self.sim.now(), self.name, Nwaiting))



yield request, self, self.sim.k, Pwait = self.sim.now() - arrive # waiting timeprint("%8.3f %s: Waited %6.3f" % (self.sim.now(), self.name, wait))yield hold, self, timeInBankyield release, self, self.sim.k

print("%8.3f %s: Completed" % (self.sim.now(), self.name))

# Model ------------------------------class BankModel(Simulation):

def run(self, aseed):""" PEM """seed(aseed)self.k = Resource(name="Counter", unitName="Karen",

qType=PriorityQ, sim=self)s = Source('Source', sim=self)self.activate(s, s.generate(number=5, interval=10.0), at=0.0)guido = Customer(name="Guido ", sim=self)self.activate(guido, guido.visit(timeInBank=12.0, P=100), at=23.0)self.simulate(until=maxTime)

# Experiment data -------------------------

maxTime = 400.0 # minutesseedVal = 787878

# Experiment ---------------------------


The resulting output is as follows. The number of customers in the queue just as each arrives is displayed in the trace.That count does not include any customer in service.

0.000 Customer00: Queue is 0 on arrival0.000 Customer00: Waited 0.00012.000 Customer00: Completed12.083 Customer01: Queue is 0 on arrival12.083 Customer01: Waited 0.00020.210 Customer02: Queue is 0 on arrival23.000 Guido : Queue is 1 on arrival24.083 Customer01: Completed24.083 Guido : Waited 1.08334.043 Customer03: Queue is 1 on arrival36.083 Guido : Completed36.083 Customer02: Waited 15.87348.083 Customer02: Completed48.083 Customer03: Waited 14.04060.083 Customer03: Completed60.661 Customer04: Queue is 0 on arrival60.661 Customer04: Waited 0.00072.661 Customer04: Completed

Reading carefully one can see that when Guido arrives Customer00 has been served and left at 12.000),Customer01 is in service and two (customers 02 and 03) are queueing. Guido has priority over those waiting



and is served before them at 24.000. When Guido leaves at 36.000, Customer02 starts service.

Priority Customers with preemption

Now we allow Guido to have preemptive priority. He will displace any customer in service when he arrives. Thatcustomer will resume when Guido finishes (unless higher priority customers intervene). It requires only a change toone line of the program, adding the argument, preemptable=True to the Resource statement in line 40.

""" bank23_OO: One counter with a priority customer with preemption """from SimPy.Simulation import (Simulation, Process, Resource, PriorityQ, hold,





c = Customer(name="Customer%02d" % (i), sim=self.sim)self.sim.activate(c, c.visit(timeInBank=12.0, P=0))t = expovariate(1.0 / interval)yield hold, self, t


def visit(self, timeInBank=0, P=0):arrive = self.sim.now() # arrival timeNwaiting = len(self.sim.k.waitQ)print("%8.3f %s: Queue is %d on arrival" %

(self.sim.now(), self.name, Nwaiting))

yield request, self, self.sim.k, Pwait = self.sim.now() - arrive # waiting timeprint("%8.3f %s: Waited %6.3f" % (self.sim.now(), self.name, wait))yield hold, self, timeInBankyield release, self, self.sim.k

print("%8.3f %s: Completed" % (self.sim.now(), self.name))


def run(self, aseed):""" PEM """seed(aseed)self.k = Resource(name="Counter", unitName="Karen",

qType=PriorityQ, preemptable=True, sim=self)s = Source('Source', sim=self)self.activate(s, s.generate(number=5, interval=10.0), at=0.0)guido = Customer(name="Guido ", sim=self)self.activate(guido, guido.visit(timeInBank=12.0, P=100), at=23.0)self.simulate(until=maxTime)



# Experiment data -------------------------


# Experiment -------- ---------------------


Though Guido arrives at the same time, 23.000, he no longer has to wait and immediately goes into service,displacing the incumbent, Customer01. That customer had already completed 23.000-12.000 = 11.000minutes of his service. When Guido finishes at 35.000, Customer01 resumes service and takes 36.000-35.000 = 1.000 minutes to finish. His total service time is the same as before (12.000 minutes).

0.000 Customer00: Queue is 0 on arrival0.000 Customer00: Waited 0.0008.634 Customer01: Queue is 0 on arrival12.000 Customer00: Completed12.000 Customer01: Waited 3.36616.016 Customer02: Queue is 0 on arrival19.882 Customer03: Queue is 1 on arrival20.246 Customer04: Queue is 2 on arrival23.000 Guido : Queue is 3 on arrival23.000 Guido : Waited 0.00035.000 Guido : Completed36.000 Customer01: Completed36.000 Customer02: Waited 19.98448.000 Customer02: Completed48.000 Customer03: Waited 28.11860.000 Customer03: Completed60.000 Customer04: Waited 39.75472.000 Customer04: Completed



3.14.4 Balking and Reneging Customers

Balking occurs when a customer refuses to join a queue if it is too long. Reneging (or, better, abandonment) occurs ifan impatient customer gives up while still waiting and before being served.

Balking Customers

Another term for a system with balking customers is one where “blocked customers” are “cleared”, termed by en-gineers a BCC system. This is very convenient analytically in queueing theory and formulae developed using thisassumption are used extensively for planning communication systems. The easiest case is when no queueing is al-lowed.

As an example let us investigate a BCC system with a single server but the waiting space is limited. We will estimatethe rate of balking when the maximum number in the queue is set to 1. On arrival into the system the customer mustfirst check to see if there is room. We will need the number of customers in the system or waiting. We could keep acount, incrementing when a customer joins the queue or, since we have a Resource, use the length of the Resource’swaitQ. Choosing the latter we test (on line 23). If there is not enough room, we balk, incrementing a class variableCustomer.numBalking at line 32 to get the total number balking during the run.

""" bank24_OO. BCC system with several counters """from SimPy.Simulation import (Simulation, Process, Resource, hold, request,





c = Customer(name="Customer%02d" % (i), sim=self.sim)self.sim.activate(c, c.visit())t = expovariate(1.0 / meanTBA)yield hold, self, t


def visit(self):arrive = self.sim.now()print("%8.4f %s: Here I am " % (self.sim.now(), self.name))if len(self.sim.k.waitQ) < maxInQueue: # the test

yield request, self, self.sim.kwait = self.sim.now() - arriveprint("%8.4f %s: Wait %6.3f" % (self.sim.now(), self.name, wait))tib = expovariate(1.0 / timeInBank)yield hold, self, tibyield release, self, self.sim.kprint("%8.4f %s: Finished " % (self.sim.now(), self.name))

else:Customer.numBalking += 1print("%8.4f %s: BALKING " % (self.sim.now(), self.name))



# Modelclass BankModel(Simulation):

def run(self, aseed):""" PEM """seed(aseed)Customer.numBalking = 0self.k = Resource(capacity=numServers,

name="Counter", unitName="Clerk", sim=self)s = Source('Source', sim=self)self.activate(s, s.generate(number=maxNumber, meanTBA=ARRint), at=0.0)self.simulate(until=maxTime)

# Experiment data -------------------------------

timeInBank = 12.0 # mean, minutesARRint = 10.0 # mean interarrival time, minutesnumServers = 1 # serversmaxInSystem = 2 # customersmaxInQueue = maxInSystem - numServers

maxNumber = 8maxTime = 4000.0 # minutestheseed = 212121

# Experiment --------------------------------------

mymodel = BankModel()mymodel.run(aseed=theseed)# Results -----------------------------------------

nb = float(Customer.numBalking)print("balking rate is %8.4f per minute" % (nb / mymodel.now()))

The resulting output for a run of this program showing balking occurring is given below:

0.0000 Customer00: Here I am0.0000 Customer00: Wait 0.0004.3077 Customer01: Here I am5.6957 Customer02: Here I am5.6957 Customer02: BALKING6.9774 Customer03: Here I am6.9774 Customer03: BALKING8.2476 Customer00: Finished8.2476 Customer01: Wait 3.940

21.1312 Customer04: Here I am22.4840 Customer01: Finished22.4840 Customer04: Wait 1.35323.0923 Customer05: Here I am23.1537 Customer06: Here I am23.1537 Customer06: BALKING36.0653 Customer04: Finished36.0653 Customer05: Wait 12.97338.4851 Customer07: Here I am53.1056 Customer05: Finished53.1056 Customer07: Wait 14.62060.3558 Customer07: Finished

balking rate is 0.0497 per minute



When Customer02 arrives, numbers 00 is already in service and 01 is waiting. There is no room so 02 balks. Bythe vagaries of exponential random numbers, 00 takes a very long time to serve (55.0607 minutes) so the first one tofind room is number 07 at 73.0765.

Reneging (or abandoning) Customers

Often in practice an impatient customer will leave the queue before being served. SimPy can model this renegingbehaviour using a compound yield statement. In such a statement there are two yield clauses. An example is:

yield (request,self,counter),(hold,self,maxWaitTime)

The first tuple of this statement is the usual yield request, asking for a unit of counter Resource. The processwill either get the unit immediately or be queued by the Resource. The second tuple is a reneging clause which hasthe same syntax as a yield hold. The requesting process will renege if the wait exceeds maxWaitTime.

There is a complication, though. The requesting PEM must discover what actually happened. Did the process get theresource or did it renege? This involves a mandatory test of self.acquired(resource). In our example, this testis in line 26.

""" bank21_OO: One counter with impatient customers """from SimPy.Simulation import (Simulation, Process, Resource, hold, request,





c = Customer(name="Customer%02d" % (i), sim=self.sim)self.sim.activate(c, c.visit(timeInBank=15.0))t = expovariate(1.0 / interval)yield hold, self, t


def visit(self, timeInBank=0):arrive = self.sim.now() # arrival timeprint("%8.3f %s: Here I am " % (self.sim.now(), self.name))

yield (request, self, self.sim.counter), (hold, self, maxWaitTime)wait = self.sim.now() - arrive # waiting timeif self.acquired(self.sim.counter):

print("%8.3f %s: Waited %6.3f" % (self.sim.now(), self.name, wait))yield hold, self, timeInBankyield release, self, self.sim.counterprint("%8.3f %s: Completed" % (self.sim.now(), self.name))

else:print("%8.3f %s: Waited %6.3f. I am off" %

(self.sim.now(), self.name, wait))



# Model ----------------------------------class BankModel(Simulation):

def run(self, aseed):""" PEM """seed(aseed)self.counter = Resource(name="Karen", sim=self)source = Source('Source', sim=self)self.activate(source,

source.generate(number=5, interval=10.0), at=0.0)self.simulate(until=maxTime)

# Experiment data -------------------------

maxTime = 400.0 # minutesmaxWaitTime = 12.0 # minutes. maximum time to waitseedVal = 989898

# Experiment ----------------------------------


0.000 Customer00: Here I am0.000 Customer00: Waited 0.0008.634 Customer01: Here I am15.000 Customer00: Completed15.000 Customer01: Waited 6.36616.016 Customer02: Here I am19.882 Customer03: Here I am20.246 Customer04: Here I am28.016 Customer02: Waited 12.000. I am off30.000 Customer01: Completed30.000 Customer03: Waited 10.11832.246 Customer04: Waited 12.000. I am off45.000 Customer03: Completed

Customer01 arrives after 00 but has only 12 minutes patience. After that time in the queue (at time 14.166) heabandons the queue to leave 02 to take his place. 03 also abandons. 04 finds an empty system and takes the serverwithout having to wait.



3.14.5 Processes

In some simulations it is valuable for one SimPy Process to interrupt another. This can only be done when the victimis “active”; that is when it has an event scheduled for it. It must be executing a yield hold statement.

A process waiting for a resource (after a yield request statement) is passive and cannot be interrupted by another.Instead the yield waituntil and yield waitevent facilities have been introduced to allow processes to waitfor conditions set by other processes.

Interrupting a Process.

Klaus goes into the bank to talk to the manager. For clarity we ignore the counters and other customers. During hisconversation his cellphone rings. When he finishes the call he continues the conversation.

In this example, call is an object of the Call Process class whose only purpose is to make the cellphone ring aftera delay, timeOfCall, an argument to its ring PEM (line 26).

klaus, a Customer, is interrupted by the call (line 29). He is in the middle of a yield hold (line 12). When heexits from that command it is as if he went into a trance when talking to the bank manager. He suddenly wakes up andmust check (line 13) to see whether has finished his conversation (if there was no call) or has been interrupted.

If self.interrupted() is False he was not interrupted and leaves the bank (line 21) normally. If it is True,he was interrupted by the call, remembers how much conversation he has left (line 14), resets the interrupt (line 15)and then deals with the call. When he finishes (line 19) he can resume the conversation, with, now we assume, athoroughly irritated bank manager v(line 20).

""" bank22_OO: An interruption by a phone call """from SimPy.Simulation import Simulation, Process, hold



def visit(self, timeInBank, onphone):print("%7.4f %s: Here I am" % (self.sim.now(), self.name))yield hold, self, timeInBankif self.interrupted():

timeleft = self.interruptLeftself.interruptReset()print("%7.4f %s: Excuse me" % (self.sim.now(), self.name))print("%7.4f %s: Hello! I'll call back" %

(self.sim.now(), self.name))yield hold, self, onphoneprint("%7.4f %s: Sorry, where were we?" %

(self.sim.now(), self.name))yield hold, self, timeleft

print("%7.4f %s: I must leave" % (self.sim.now(), self.name))

class Call(Process):""" Cellphone call arrives and interrupts """

def ring(self, klaus, timeOfCall):yield hold, self, timeOfCallprint("%7.4f Ringgg!" % (self.sim.now()))self.interrupt(klaus)



# Model -----------------------------------

class BankModel(Simulation):def run(self):

""" PEM """klaus = Customer(name="Klaus", sim=self)self.activate(klaus, klaus.visit(timeInBank, onphone))call = Call(sim=self)self.activate(call, call.ring(klaus, timeOfCall))self.simulate(until=maxTime)

# Experiment data -------------------------

timeInBank = 20.0timeOfCall = 9.0onphone = 3.0maxTime = 100.0

# Experiment -----------------------------mymodel = BankModel()mymodel.run()

0.0000 Klaus: Here I am9.0000 Ringgg!9.0000 Klaus: Excuse me9.0000 Klaus: Hello! I'll call back

12.0000 Klaus: Sorry, where were we?23.0000 Klaus: I must leave

As this has no random numbers the results are reasonably clear: the interrupting call occurs at 9.0. It takes klaus3 minutes to listen to the message and he resumes the conversation with the bank manager at 12.0. His total time ofconversation is 9.0 + 11.0 = 20.0 minutes as it would have been if the interrupt had not occurred.

waituntil the Bank door opens

Customers arrive at random, some of them getting to the bank before the door is opened by a doorman. They waitfor the door to be opened and then rush in and queue to be served. The door is modeled by an attribute door ofBankModel.

This model uses the waituntil yield command. In the program listing the door is initially closed (line 58) and amethod to test if it is open is defined at line 54.

The Doorman class is defined starting at line 7 and the single doorman is created and activated at at lines 59 and 60.The doorman waits for an average 10 minutes (line 11) and then opens the door.

The Customer class is defined at 24 and a new customer prints out Here I am on arrival. If the door is still closed,he adds but the door is shut and settles down to wait (line 35), using the yield waituntil command.When the door is opened by the doorman the dooropen state is changed and the customer (and all others waiting forthe door) proceed. A customer arriving when the door is open will not be delayed.

"""bank14_OO: *waituntil* the Bank door opens"""from SimPy.Simulation import (Simulation, Process, Resource, hold, waituntil,





class Doorman(Process):""" Doorman opens the door"""

def openthedoor(self):""" He will open the door when he arrives"""yield hold, self, expovariate(1.0 / 10.0)self.sim.door = 'Open'print("%7.4f Doorman: Ladies and "

"Gentlemen! You may all enter." % (self.sim.now()))



c = Customer(name="Customer%02d" % (i), sim=self.sim)self.sim.activate(c, c.visit(timeInBank=12.0))yield hold, self, expovariate(rate)


def visit(self, timeInBank=10):arrive = self.sim.now()

if self.sim.dooropen():msg = ' and the door is open.'


print("%7.4f %s: Here I am%s" % (self.sim.now(), self.name, msg))

yield waituntil, self, self.sim.dooropen

print("%7.4f %s: I can go in!" % (self.sim.now(), self.name))wait = self.sim.now() - arriveprint("%7.4f %s: Waited %6.3f" % (self.sim.now(), self.name, wait))

yield request, self, self.sim.countertib = expovariate(1.0 / timeInBank)yield hold, self, tibyield release, self, self.sim.counter

print("%7.4f %s: Finished " % (self.sim.now(), self.name))

# Model ----------------------------------

class BankModel(Simulation):def dooropen(self):

return self.door == 'Open'

def run(self, aseed):



""" PEM """seed(aseed)self.counter = Resource(capacity=1, name="Clerk", sim=self)self.door = 'Shut'doorman = Doorman(sim=self)self.activate(doorman, doorman.openthedoor())source = Source(sim=self)self.activate(source,

source.generate(number=5, rate=0.1), at=0.0)self.simulate(until=400.0)

# Experiment data -------------------------


# Experiment ----------------------------------



0.0000 Customer00: Here I am but the door is shut.1.1489 Doorman: Ladies and Gentlemen! You may all enter.1.1489 Customer00: I can go in!1.1489 Customer00: Waited 1.1496.5691 Customer00: Finished8.3438 Customer01: Here I am and the door is open.8.3438 Customer01: I can go in!8.3438 Customer01: Waited 0.000

15.5704 Customer02: Here I am and the door is open.15.5704 Customer02: I can go in!15.5704 Customer02: Waited 0.00021.2664 Customer03: Here I am and the door is open.21.2664 Customer03: I can go in!21.2664 Customer03: Waited 0.00021.9473 Customer04: Here I am and the door is open.21.9473 Customer04: I can go in!21.9473 Customer04: Waited 0.00027.6401 Customer01: Finished56.5248 Customer02: Finished57.3640 Customer03: Finished77.3587 Customer04: Finished

Wait for the doorman to give a signal: waitevent

Customers arrive at random, some of them getting to the bank before the door is open. This is controlled by anautomatic machine called the doorman which opens the door only at intervals of 30 minutes (it is a very secure bank).The customers wait for the door to be opened and all those waiting enter and proceed to the counter. The door is closedbehind them.

This model uses the yield waitevent command which requires a SimEvent attribute for BankModel to bedefined (line 56). The Doorman class is defined at line 7 and the doorman is created and activated at at labels 56 and57. The doorman waits for a fixed time (label 12) and then tells the customers that the door is open. This is achieved



on line 13 by signalling the dooropen event.

The Customer class is defined at 24 and in its PEM, when a customer arrives, he prints out Here I am. If thedoor is still closed, he adds “but the door is shut‘ and settles down to wait for the door to be opened using the yieldwaitevent command (line 34). When the door is opened by the doorman (that is, he sends the dooropen.signal() the customer and any others waiting may proceed.

""" bank13_OO: Wait for the doorman to give a signal: *waitevent*"""from SimPy.Simulation import (Simulation, Process, Resource, SimEvent, hold,

request, release, waitevent)from random import *


class Doorman(Process):""" Doorman opens the door"""

def openthedoor(self):""" He will opens the door at fixed intervals"""for i in range(5):

yield hold, self, 30.0self.sim.dooropen.signal()print("%7.4f You may enter" % (self.sim.now()))





def visit(self, timeInBank=10):arrive = self.sim.now()

if self.sim.dooropen.occurred:msg = '.'


print("%7.4f %s: Here I am%s" % (self.sim.now(), self.name, msg))yield waitevent, self, self.sim.dooropen

print("%7.4f %s: The door is open!" % (self.sim.now(), self.name))

wait = self.sim.now() - arriveprint("%7.4f %s: Waited %6.3f" % (self.sim.now(), self.name, wait))

yield request, self, self.sim.countertib = expovariate(1.0 / timeInBank)yield hold, self, tibyield release, self, self.sim.counter




# Model ----------------------------------


""" PEM """seed(aseed)self.dooropen = SimEvent("Door Open", sim=self)self.counter = Resource(1, name="Clerk", sim=self)doorman = Doorman(sim=self)self.activate(doorman, doorman.openthedoor())source = Source(sim=self)self.activate(source,

source.generate(number=5, rate=0.1), at=0.0)self.simulate(until=maxTime)

# Experiment data -------------------------


# Experiment ----------------------------------



0.0000 Customer00: Here I am but the door is shut.13.6767 Customer01: Here I am but the door is shut.13.9068 Customer02: Here I am but the door is shut.30.0000 You may enter30.0000 Customer02: The door is open!30.0000 Customer02: Waited 16.09330.0000 Customer01: The door is open!30.0000 Customer01: Waited 16.32330.0000 Customer00: The door is open!30.0000 Customer00: Waited 30.00034.0411 Customer03: Here I am but the door is shut.40.8095 Customer04: Here I am but the door is shut.55.4721 Customer02: Finished57.2363 Customer01: Finished60.0000 You may enter60.0000 Customer04: The door is open!60.0000 Customer04: Waited 19.19060.0000 Customer03: The door is open!60.0000 Customer03: Waited 25.95977.0409 Customer00: Finished90.0000 You may enter104.8327 Customer04: Finished118.4142 Customer03: Finished120.0000 You may enter150.0000 You may enter



3.14.6 Monitors

Monitors (and Tallys) are used to track and record values in a simulation. They store a list of [time,value] pairs, onepair being added whenever the observe method is called. A particularly useful characteristic is that they continue toexist after the simulation has been completed. Thus further analysis of the results can be carried out.

Monitors have a set of simple statistical methods such as mean and var to calculate the average and variance of theobserved values – useful in estimating the mean delay, for example.

They also have the timeAverage method that calculates the time-weighted average of the recorded values. Itdetermines the total area under the time~value graph and divides by the total time. This is useful for estimating theaverage number of customers in the bank, for example. There is an important caveat in using this method. To estimatethe correct time average you must certainly observe the value (say the number of customers in the system) wheneverit changes (as well as at any other time you wish) but, and this is important, observing the new value. The old valuewas recorded earlier. In practice this means that if we wish to observe a changing value, n, using the Monitor, Mon,we must keep to the the following pattern:

n = n+1Mon.observe(n,self.sim.now())

Thus you make the change (not only increases) and then observe the new value. Of course the simulation time now()has not changed between the two statements.

Plotting a Histogram of Monitor results

A Monitor can construct a histogram from its data using the histogram method. In this model we monitor the timein the system for the customers. This is calculated for each customer in line 29, using the arrival time saved in line 19.We create the Monitor attribute of BankModel, Mon, at line 39 and the times are observed at line 30.

The histogram is constructed from the Monitor, after the simulation has finished, at line 58. The SimPy SimPlotpackage allows simple plotting of results from simulations. Here we use the SimPlot plotHistogram method. Theplotting routines appear in lines 60-64. The plotHistogram call is in line 61.

"""bank17_OO: Plotting a Histogram of Monitor results"""from SimPy.Simulation import (Simulation, Process, Resource, Monitor, hold,

request, release)from SimPy.SimPlot import *from random import expovariate, seed






def visit(self, timeInBank):arrive = self.sim.now()



# print("%8.4f %s: Arrived "%(now(), self.name))

yield request, self, self.sim.counter# print("%8.4f %s: Got counter "%(now(), self.name))tib = expovariate(1.0 / timeInBank)yield hold, self, tibyield release, self, self.sim.counter

# print("%8.4f %s: Finished " % (now(), self.name))t = self.sim.now() - arriveself.sim.Mon.observe(t)

# Model ----------------------------------


""" PEM """seed(aseed)self.counter = Resource(1, name="Clerk", sim=self)self.Mon = Monitor('Time in the Bank', sim=self)source = Source(sim=self)self.activate(source,


# Experiment data -------------------------


N = 0seedVal = 393939

# Experiment -----------------------------

modl = BankModel()modl.run(aseed=seedVal)

# Output ----------------------------------Histo = modl.Mon.histogram(low=0.0, high=200.0, nbins=20)

plt = SimPlot()plt.plotHistogram(Histo, xlab='Time (min)',

title="Time in the Bank",color="red", width=2)

plt.mainloop()

Monitoring a Resource

Now consider observing the number of customers waiting or executing in a Resource. Because of the need toobserve the value after the change but at the same simulation instant, it is impossible to use the length of theResource’s waitQ directly with a Monitor defined outside the Resource. Instead Resources can be set up with built-inMonitors.

Here is an example using a Monitored Resource. We intend to observe the average number waiting and active in thecounter resource. counter is defined at line 35 as a BankModel attribute and we have set monitored=True.



This establishes two Monitors: waitMon, to record changes in the numbers waiting and actMon to record changesin the numbers active in the counter. We need make no further change to the operation of the program as monitoringis then automatic. No observe calls are necessary.

After completion of the run method, we calculate the timeAverage of both waitMon and actMon (lines 53-54).These can then be printed at the end of the program (line 55).

"""bank15_OO: Monitoring a Resource"""from SimPy.Simulation import (Simulation, Process, Resource, hold, request,

release)from random import *




c = Customer(name="Customer%02d" % (i), sim=self.sim)self.sim.activate(c, c.visit(

timeInBank=12.0, counter=self.sim.counter))yield hold, self, expovariate(rate)


def visit(self, timeInBank, counter):arrive = self.sim.now()print("%8.4f %s: Arrived " % (self.sim.now(), self.name))

yield request, self, counterprint("%8.4f %s: Got counter " % (self.sim.now(), self.name))tib = expovariate(1.0 / timeInBank)yield hold, self, tibyield release, self, counter


# Model ----------------------------------


""" PEM """seed(aseed)self.counter = Resource(capacity=1, name="Clerk",

monitored=True, sim=self)source = Source(sim=self)self.activate(source,


return

# Experiment data -------------------------




# Experiment ----------------------------------

modl = BankModel()modl.run(aseed=seedVal)

nrwaiting = modl.counter.waitMon.timeAverage()nractive = modl.counter.actMon.timeAverage()print('Average waiting = %6.4f\nAverage active = %6.4f\n' %

(nrwaiting, nractive))

Plotting from Resource Monitors

Like all Monitors, waitMon and actMon in a monitored Resource contain information that enables us to graph theoutput. Alternative plotting packages can be used; here we use the simple SimPy.SimPlot package just to graphthe number of customers waiting for the counter. The program is a simple modification of the one that uses a monitoredResource.

The SimPlot package is imported at line 3. No major changes are made to the main part of the program except that Icommented out the print statements. The changes occur in the run method from lines 38 to 39. The simulation nowgenerates and processes 20 customers (line 39). The Monitors of the counter Resource attribute still exist when thesimulation has terminated.

The additional plotting actions take place in lines 54 to 57. Line 55-56 construct a step plot and graphs the numberin the waiting queue as a function of time. waitMon is primarily a list of [time,value] pairs which the plotStepmethod of the SimPlot object, plt uses without change. On running the program the graph is plotted; the user has toterminate the plotting mainloop on the screen.

"""bank16_OO: Plotting from Resource Monitors"""from SimPy.Simulation import (Simulation, Process, Resource, hold, request,

release)from SimPy.SimPlot import *from random import expovariate, seed






def visit(self, timeInBank):arrive = self.sim.now()# print("%8.4f %s: Arrived " % (now(), self.name))



yield request, self, self.sim.counter# print("%8.4f %s: Got counter " % (now(), self.name))tib = expovariate(1.0 / timeInBank)yield hold, self, tibyield release, self, self.sim.counter

# print("%8.4f %s: Finished " % (now(), self.name))

# Model -----------------------------------


""" PEM """seed(aseed)self.counter = Resource(1, name="Clerk", monitored=True, sim=self)source = Source(sim=self)self.activate(source,


# Experiment data -------------------------


# Experiment -----------------------------------


# Output ---------------------------------------

plt = SimPlot()plt.plotStep(mymodel.counter.waitMon,

color="red", width=2)plt.mainloop()


I thank Klaus Muller, Bob Helmbold, Mukhlis Matti and the other developers and users of SimPy for improving thisdocument by sending their comments. I would be grateful for any further corrections or suggestions. Please send themto: vignaux at users.sourceforge.net.

3.14.8 References


• SimPy homepage: https://github.com/SimPyClassic/SimPyClassic

• The Bank:





3.15 SimPy Course on the Web

An outstanding tutorial on SimPy by Prof. Norm Matloff (U. of California at Davis) can be found on the Web: ADiscrete-Event Simulation Course Based on the SimPy Language.

This course material has been developed by Prof. Matloff in his SimPy courses at U. of California. It is being evolvedfurther, so keep going back to this great teaching material!

3.15. SimPy Course on the Web 319





CHAPTER 4

Interfacing to External Packages

4.1 Publication-quality plot production with matplotlib

This document deals with producing production-quality plots from SimPy simulation output using the matplotliblibrary. matplotlib is known to work on Linux, Unix, MS Windows and OS X platforms. This library is not part of theSimPy distribution and has to be downloaded and installed separately.

Simulation programs normally produce large quantities of output which needs to be visualized, e.g. by plotting. Theseplots can help with aggregating data, e.g. for detecting trends over time, frequency distributions or determining thewarm-up period of a simulation model experiment.

SimPy’s SimPlot plotting package is an easy to use, out-of-the-box capability which can produce a full range of plotgraphs on the screen and in PostScript format. After installing SimPy, it can be used without installing any othersoftware. It is tightly integrated with SimPy, e.g. its Monitor data collection class.

The SimPlot library is not intended to produce publication-quality plots. If you want to publish your plots in a reportor on the web, consider using an external plotting library which can be called rom Python.

4.1.1 About matplotlib

A very popular plotting library for Python is matplotlib. Its capabilities far exceed those of SimPy’s SimPlot. This ishow matplotlib is described on its home page:

“matplotlib is a python 2D plotting library which produces publication quality figures in a variety ofhardcopy formats and interactive environments across platforms. matplotlib can be used in python scripts,the python and ipython shell (a la matlab or mathematica), web application servers, and six graphicaluser interface toolkits.”

The matplotlib screenshots (with Python code) at https://matplotlib.org/gallery/index.html show the great range ofquality displays the library can produce with little coding. For the investment in time in downloading, installing andlearning matplotlib, the SimPy user is rewarded with a powerful plotting capability.

321

https://matplotlib.org/gallery/index.html


Downloading matplotlib

Extensive installation instructions are provided at https://matplotlib.org/users/installing.html

matplotlib input data

matplotlib takes separate sequences (lists, tuples, arrays) for x- and y-values. SimPlot, on the other hand, plots Monitorinstances, i.e., lists of [x,y] lists.

This difference in data structures is easy to overcome in SimPy by using the Monitor functions yseries (returningthe list of y-data) and tseries (returning the list of time- or x-data).

An example from the Bank Tutorial

As an example of how to use matplotlib with SimPy, a modified version of bank12.py from the Bank Tutorial is usedhere. It produces a line plot of the counter’s queue length and a histogram of the customer wait times:

""" Based on bank12.py in Bank Tutorial.This program uses matplotlib. It produces two plots:- Queue length over time- Histogram of queue length"""import SimPy.Simulation as Simulationimport pylab as pylfrom random import Random

# Model componentsclass Source(Simulation.Process):

""" Source generates customers randomly"""

def __init__(self, seed=333):Simulation.Process.__init__(self)self.SEED = seed


c = Customer(name="Customer%02d" % (i,))Simulation.activate(c, c.visit(timeInBank=12.0))t = rv.expovariate(1.0 / interval)yield Simulation.hold, self, t

class Customer(Simulation.Process):""" Customer arrives, is served and leaves """


def visit(self, timeInBank=0):arrive = Simulation.now()yield Simulation.request, self, counterwait = Simulation.now() - arrivewate.observe(y=wait)

322 Chapter 4. Interfacing to External Packages

https://matplotlib.org/users/installing.html


tib = counterRV.expovariate(1.0 / timeInBank)yield Simulation.hold, self, tibyield Simulation.release, self, counter

class Observer(Simulation.Process):def __init__(self):

Simulation.Process.__init__(self)

def observe(self):while True:

yield Simulation.hold, self, 5qu.observe(y=len(counter.waitQ))

# Modeldef model(counterseed=3939393):

global counter, counterRV, waitMonitorcounter = Simulation.Resource(name="Clerk", capacity=1)counterRV = Random(counterseed)waitMonitor = Simulation.Monitor()Simulation.initialize()sourceseed = 1133source = Source(seed=sourceseed)Simulation.activate(source, source.generate(100, 10.0))ob = Observer()Simulation.activate(ob, ob.observe())Simulation.simulate(until=2000.0)

qu = Simulation.Monitor(name="Queue length")wate = Simulation.Monitor(name="Wait time")# Experiment datasourceSeed = 333# Experimentmodel()# Outputpyl.figure(figsize=(5.5, 4))pyl.plot(qu.tseries(), qu.yseries())pyl.title("Bank12: queue length over time",

fontsize=12, fontweight="bold")pyl.xlabel("time", fontsize=9, fontweight="bold")pyl.ylabel("queue length before counter", fontsize=9, fontweight="bold")pyl.grid(True)pyl.show()pyl.savefig("./bank12.png")print("Saved graph in current directory as bank12.png")

pyl.clf()n, bins, patches = pyl.hist(qu.yseries(), 10, normed=True)pyl.title("Bank12: Frequency of counter queue length",

fontsize=12, fontweight="bold")pyl.xlabel("queuelength", fontsize=9, fontweight="bold")pyl.ylabel("frequency", fontsize=9, fontweight="bold")pyl.grid(True)pyl.xlim(0, 30)pyl.show()pyl.savefig("./bank12histo.png")

4.1. Publication-quality plot production with matplotlib 323


print("Saved graph in current directory as bank12histo.png")

Here is the explanation of this program:

Line number and explanation

07 Imports the matplotlib pylab module (this import form is needed to avoid namespace clashes with SimPy).

75 Sets the size of the figures following to a width of 5.5 and a height of 4 inches.

76 Plots the series of queue-length values (qu.yseries()) over their observation times series (qu.tseries()).

77 Sets the figure title, its font size, and its font weight.

79 Sets the x-axis label, its font size, and its font weight.

80 Sets the y-axis label, its font size, and its font weight.

81 Gives the graph a grid.

83 Saves the plot under the given name.

86 Clears the current figure (e.g., resets the axes values from the previous plot).

87 Makes a histogram of the queue-length series (qu.series()) with 10 bins. The normed parameter makes the fre-quency counts relative to 1.

88 Sets the title etc.

90 Sets the x-axis label etc.

91 Sets the y-axis label etc.

92 Gives the graph a grid.

93 Limits the x-axis to the range[0..30].

95 Saves the plot under the given name.

Running the program above results in two PNG files. The first (bank12.png) shows the queue length over time:



The second output file (bank12histo.png) is a histogram of the customer queue length at the counter:

4.1. Publication-quality plot production with matplotlib 325


4.1.2 Conclusion

The small example above already shows the power, flexibility and quality of the graphics capabilities provided bymatplotlib. Almost anything (fonts, graph sizes, line types, number of series in one plot, number of subplots in a plot,. . . ) is under user control by setting parameters or calling functions. Admittedly, it initially takes a lot of reading inthe extensive documentation and some experimentation, but the results are definitely worth the effort!

4.2 Running SimPy on Multiple Processors with Parallel Python

Contents

• Running SimPy on Multiple Processors with Parallel Python

– Introduction

– Examples

4.2.1 Introduction

With SimPy 2.0, you can easily increase the performance of your simulation by using Parallel Python if you have alarger number of independent processors (multiple CPUs or cores). Parallel Python can distribute the execution ofyour SimPy processes to all cores of your CPU and even to other computers. You should read the PP documentationfor further information on how this works.

Please, note that Parallel Python is not included in the SimPy distribution and needs to be downloaded<https://www.parallelpython.com/> and installed separately.

4.2.2 Examples

Example #1

The files PPExample.txt and PPExampleProcess.txt contain a small example with several car processes. It is important,that processes etc.are not defined in the file that starts the PP job server and executes the jobs, since ppserver.submit() only takes functions defined in the same file and module names to import, but no classes.

PPExample.py:

"""Example for SimPy with Parallel Python."""

import PPExampleProcessfrom SimPy.Simulation import *import pp

def runSimulation(jobNum, numCars):sim = SimPy.Simulation.Simulation()cars = []for i in range(numCars):

car = PPExampleProcess.Car(sim, i * jobNum + i)sim.activate(car, car.run(), at = 0)


https://www.parallelpython.com/

../../../_static/PPExample.txt

../../../_static/PPExampleProcess.txt


cars.append(car)sim.simulate(until = 30)

server = pp.Server(ppservers = ())for i in range(4):

job = server.submit(runSimulation,(i, 2),(),('SimPy.Simulation', 'PPExampleProcess'))

job()

PPExampleProcess.py:


class Car(Process):

def __init__(self, sim, id):Process.__init__(self, sim = sim)self.id = id

def run(self):while True:

yield hold, self, 10print 'Car #%i at t = %i' % (self.id, self.sim.now())

The simulated process in this case is a simple car, that holds for ten steps and then prints the current simulation time.Obviously, each car process is independent from the other ones. Thus if we want to simulate a great number of cars,we can easily distribute the processes to many CPU cores and others computers in our network.

PP pickles everything it sends to other cores/computers. Since SimPy is currently not pickleable, you cannot submitSimulation.simulate() to the PPServer. In this example runSimulation is defined and submitted to theserver. The code within it will be executed on each core/computer. In the example we create four simulation jobs withtwo cars for each job.

Run PPExample.py to execute the example.

Example #2

Files simulator.txt and processes.txt contain an example that simulates refrigerators in single-thread and in parallelsimulation mode.

Class Simulator simulates a number of fridges and gets the resulting data.

Class ParallelSimulator simulates a number of fridges and gets the resulting data. A number of jobs will becreated that use all available CPU cores or even other computers.

To use clustering, ParallelPython needs to be installed on all computers and the server demon “ppserver.py” must bestarted. The list of the servers’ IP addresses must then be passed to the constructor of this class.

Run simulator.py to execute this example.

File simulator.py:

# coding=utf8"""The fridge simulation

4.2. Running SimPy on Multiple Processors with Parallel Python 327

../../../_static/simulator.txt

../../../_static/processes.txt


@author: Stefan Scherfke@contact: stefan.scherfke at uni-oldenburg.de"""

from time import clockimport logging

from SimPy.Simulation import Simulation, activate, initialize, simulateimport pp

from processes import Fridge, FridgeObserver

log = logging.getLogger('Simulator')

class Simulator(object):"""This class simulates a number of fridges and gets the resulting data."""

def __init__(self, numFridges, tau, aggSteps, duration):"""Setup the simulation with the specified number of fridges.

Tau specifies the simulation step for each frige. Furthermore theobserver will collect data each tau. Collected datawill be aggregated at the end of each aggSteps simulation steps.

@param numFridges: The number of simulated fridges@type numFridges: unsigned int@param tau: simulation step size for collecting data and simulating

the fridge@type tau: float@param aggSteps: Collected data will be aggregated each aggSteps

simulation steps. Signals interval will betau * aggSteps

@type aggSteps: unsigned int@param duration: Duration of the simulation in hours@type duration: unsigned int"""log.info('Initializing simulator ...')self.simEnd = durationself.sim = Simulation()

fridgeProperties = {'tau': tau}self._fridges = []for i in range(numFridges):

fridge = Fridge(self.sim, **fridgeProperties)self._fridges.append(fridge)

self._observer = FridgeObserver(self.sim, self._fridges, tau, aggSteps)

def simulate(self):"""Initialize the system, start the simulation and return the collecteddata.

@return: The fridgerators consumption after each aggregation"""



log.info('Running simulation ...')self.sim.initialize()for fridge in self._fridges:

self.sim.activate(fridge, fridge.run(), at = 0)self.sim.activate(self._observer, self._observer.run(), at = 0)self.sim.simulate(until = self.simEnd)

log.info('Simulation run finished.')return self._observer.getData()

class ParallelSimulator(object):"""This class simulates a number of fridges and gets the resulting data.Unlike simulator, a number of jobs will be created that use all availaleCPU cores or even other computers.

To use clustering, ParallelPython needs to be installed on all computersand the server demon "ppserver.py" must be started. The list of the server'sIPs must then be passed to the constructor of this class."""

def __init__(self, numFridges, tau, aggSteps, duration,jobSize = 100, servers = ()):

"""Setup the simulation with the specified number of fridges. It will besplit up in several parallel jobs, each with the specified number ofjobs.

Tau specifies the simulation step for each frige. Furthermore theobserver will collect data each tau. Collected datawill be aggregated at the end of each aggSteps simulation steps.

@param numFridges: The number of simulated fridges@type numFridges: unsigned int@param tau: simulation step size for collecting data and simulatingthe fridge@type tau: float@param aggSteps: Collected data will be aggregated each aggStepssimulation steps. Signals interval will betau * aggSteps@type aggSteps: unsigned int@param duration: Duration of the simulation@type duration: unsigned int@param jobSize: The number of friges per job, defaults to 100.@type jobSize: unsigned int@param servers: A list of IPs from on which the simulation shall be

executed. Defaults to "()" (use only SMP)@type servers: tuple of string"""log.info('Initializing prallel simulation ...')

self._jobSize = jobSizeself._servers = serversself._numFridges = numFridgesself._tau = tauself._aggSteps = aggStepsself.simEnd = duration



def simulate(self):"""Create some simulation jobs, run them and retrieve their results.

@return: The fridgerators consumption after each aggregation"""log.info('Running parallel simulation ...')oldLevel = log.getEffectiveLevel() # pp changes the log level :(jobServer = pp.Server(ppservers = self._servers)

# Start the jobsremainingFridges = self._numFridgesjobs = []while remainingFridges > 0:

jobs.append(jobServer.submit(self.runSimulation,(min(self._jobSize, remainingFridges),),(),("logging", "SimPy.Simulation", "processes")))

remainingFridges -= self._jobSizelog.info('Number of jobs for simulation: %d' % len(jobs))

# Add each job's datapSum = [0] * int((60 / self._aggSteps) * self.simEnd)for job in jobs:

data = job()for i in range(len(data)):

pSum[i] += data[i]for s in pSum:

s /= len(jobs)

log.setLevel(oldLevel)log.info('Parallel simulation finished.')return pSum

def runSimulation(self, numFridges):"""Create a job with the specified number of fridges and controllers andone observer. Simulate this and return the results.

@param numFridges: The number of fridges to use for this job@type numFridges: unsigned int@return: A list with the aggregated fridge consumption"""sim = SimPy.Simulation.Simulation()sim.initialize()

fridgeProperties = {'tau': self._tau}fridges = []for i in range(numFridges):

fridge = processes.Fridge(sim, **fridgeProperties)fridges.append(fridge)sim.activate(fridge, fridge.run(), at = 0)

observer = processes.FridgeObserver(sim,fridges, self._tau, self._aggSteps)

sim.activate(observer, observer.run(), at = 0)

sim.simulate(until = self.simEnd)



return observer.getData()

if __name__ == '__main__':logging.basicConfig(

level = logging.INFO,format = '%(asctime)s %(levelname)8s: %(name)s: %(message)s')

numFridges = 5000tau = 1./60aggStep = 15duration = 4 + tau

sim = Simulator(numFridges, tau, aggStep, duration)data = sim.simulate()log.info('Results: ' + str(data))

servers = ()sim = ParallelSimulator(numFridges, tau, aggStep, duration, 100, servers)data = sim.simulate()log.info('Results: ' + str(data))

File process.py:

# coding=utf8"""This file contains classes for simulating, controlling and observing a fridge.

@author: Stefan Scherfke@contact: stefan.scherfke at uni-oldenburg.de"""

from math import expimport loggingimport random

from SimPy.Simulation import Process, Simulation, \activate, hold, initialize, now, simulate

log = logging.getLogger('Processes')

class Fridge(Process):"""This class represents a simulated fridge.

It's temperature T for and equidistant series of time steps is computed by$T_{i+1} = \epsilon \cdot T_i + (1 - \epsilon) \cdot \left(T^O - \eta\cdot \frac{q_i}{A}\right)$ with $\epsilon = e^{-\frac{\tau A}{m_c}}$."""

def __init__(self, sim, T_O = 20.0, A = 3.21, m_c = 15.97, tau = 1.0/60,eta = 3.0, q_i = 0.0, q_max = 70.0,T_i = 5.0, T_range = [5.0, 8.0], noise = False):

"""Init all required variables.

@param sim: The SimPy simulation this process belongs to@type sim: SimPy.Simulation



@param T_O: Outside temperature@param A: Insulation@param m_c: Thermal mass/thermal storage capacity@param tau: Time span between t_i and t_{i+1}@param eta: Efficiency of the cooling device@param q_i: Initial/current electrical power@param q_max: Power required during cool-down@param T_i: Initial/current temperature@param T_range: Allowed range for T_i@param noise: Add noise to the fridge's parameters, if True@type noise: bool"""Process.__init__(self, sim = sim)self.T_O = T_Oself.A = Aself.m_c = random.normalvariate(20, 4.5) if noise else m_cself.tau = tauself.eta = etaself.q_i = q_iself.q_max = q_maxself.T_i = random.uniform(T_range[0], T_range[1]) if noise else T_iself.T_range = T_range

def run(self):"""Calculate the fridge's temperature for the current time step."""while True:

epsilon = exp(-(self.tau * self.A) / self.m_c)self.T_i = epsilon * self.T_i + (1 - epsilon) \

* (self.T_O - self.eta * (self.q_i / self.A))if self.T_i >= self.T_range[1]:

self.q_i = self.q_max # Cool downelif self.T_i <= self.T_range[0]:

self.q_i = 0.0 # Stop coolinglog.debug('T_i: %2.2f°C at %.2f' % (self.T_i, self.sim.now()))yield hold, self, self.tau

def coolDown(self):"""Start cooling down now!"""self.q_i = self.q_max

class FridgeObserver(Process):"""This process observes the temperature and power consumption of a set offridges."""

def __init__(self, sim, fridges, tau, aggSteps):"""Init the observer.

@param sim: The SimPy simulation this process belongs to@type sim: SimPy.Simulation@param fridges: A list of fridges to be observed



@type fridges: tuple of Fridge@param tau: Time interval for observations@type tau: float@param aggSteps: Specifies after how many timesteps tau the collected

data is aggregated and stored.@type aggSteps: int"""Process.__init__(self, sim = sim)self._fridges = fridgesself._tau = tauself._aggSteps = aggStepsself._data = []

def run(self):"""Start observation"""aggSteps = 0consumption = 0lastProgUpdate = 0while True:

prog = self.sim.now() * 100 / self.sim._endtimeif int(prog) > lastProgUpdate:

log.info('Progress: %d%%' % prog)lastProgUpdate = prog

if (aggSteps >= self._aggSteps):log.debug('Aggregating at %.2f' % self.sim.now())self._data.append(consumption/self._aggSteps)consumption = 0aggSteps = 0

for fridge in self._fridges:consumption += fridge.q_i

aggSteps += 1yield hold, self, self._tau

def getData(self):"""Return the collected data

@return: a list with the collected data"""return self._data

if __name__ == '__main__':logging.basicConfig(

level = logging.DEBUG,format = '%(levelname)-8s %(asctime)s %(name)s: %(message)s')

tau = 1./60 # Step size 1minaggSteps = 15 # Aggregate consumption in 15min blocksparams = {'tau': tau}

sim = Simulation()

fridge = Fridge(sim, **params)observer = FridgeObserver(sim, [fridge], tau, aggSteps)



sim.activate(fridge, fridge.run(), at = 0)sim.activate(observer, observer.run(), at = 0)sim.simulate(until = 4 + tau)print observer.getData()


CHAPTER 5

SimPy Classic Tools

This section contains descriptions of tools written for SimPy Classic which help with its use.

5.1 SimPlot Manual

Authors Klaus Muller <[email protected]>

SimPy Classic Release 2.3.3



Date December 2011


Contents

• SimPlot Manual

– Introduction

– Plotting with SimPlot - an overview

– Simple plotting API

– Advanced plotting API

– Colours in SimPlot

The SimPlot plotting library has been developed for SimPy users so that they can produce, view and print simpleplots, without having to download and install any other software package.

This manual is aimed at the SimPy applications programmer. It describes the capabilities of SimPlot and its program-ming interface.

335




There are several more elaborate Open Source plotting packages downloadable from the Internet which can be usedfrom Python and therefore from SimPy. SimPlot is the “quick and dirty”, out-of-the-box plotting package for SimPy.If you need more complex plots or publication-quality graphics, consider using e.g. Matplotlib ( <https://matplotlib.org/>_ ).

5.1.1 Introduction

SimPlot is a basic plotting package based on Tk/Tkinter and designed for use with SimPy. It has been developed froman Open Source plotting package published in John E. Grayson’s excellent book ‘Python and Tkinter Programming’(ISBN 1-884777-81-3) which in turn was derived from Konrad Hinsen’s graph widget for NumPy (Numerical Python).

SimPlot provides for the generation, viewing and Postscript output of a variety of plot types by SimPy programs. Thedata series of SimPy Monitor instances can be plotted automatically.

SimPlot requires Tk/Tkinter to be installed (for all major operating systems on which Python runs, the Python installa-tion already includes the installation of Tk/Tkinter, so no additional download or installation is required). Test whetherTk/Tkinter is installed by running ‘import Tkinter’ on the Python interpreter command line. If you don’t get an errormessage, it is.

To write SimPlot-based programs, only a very rudimentary understanding of Tk/Tkinter is required. This manual doesnot attempt to teach Tk/Tkinter!

5.1.2 Plotting with SimPlot - an overview

Plot types and capabilities

A simple plot program using SimPlot basically looks like:

# Prog1.pyfrom SimPy.SimPlot import *plt = SimPlot()plt.plotLine([[0, 0], [1, 1], [2, 4], [3, 9]])plt.mainloop()

When running this program, the resulting output on Windows is (the outside frame will look different on other plat-forms):

The program shows the basic structure of any program using SimPlot:

336 Chapter 5. SimPy Classic Tools

https://matplotlib.org/

https://matplotlib.org/


• Line 2 imports the plotting module,

• Line 3 creates and instance of the plotting class,

• Line 4 plots a line in an x/y coordinate system,

• Line 5 starts the main loop of Tk.

The frame also shows a ‘File’ menu item (when clicked, it offers a submenu item ‘Postscript’ which allows saving theplot to a Postscript file.

Method plotline has many name parameters with default values. Here is an example showing some of them (they willall be discussed further down in this manual:

# Prog2.pyfrom SimPy.SimPlot import *plt = SimPlot()plt.plotLine([[0, 0], [1, 1], [2, 4], [3, 9]], title="This is prettier",

color="red", width=2, smooth=True)plt.mainloop()

This produces the following plot (the outside frame is not shown):

The plot now has a title and the curve is red, wider and smooth.

In addition to line plots, there are three other plot-types available in SimPlot, namely stepped line plots, bar charts,and scatter diagrams.

Here are examples of each. First, the stepped line plot:

# Prog3.pyfrom SimPy.SimPlot import *plt = SimPlot()plt.plotStep([[0, 0], [1, 1], [2, 4], [3, 9]],

color="red", width=2)plt.mainloop()

which produces:

5.1. SimPlot Manual 337


A bar chart program:

# Prog4.pyfrom SimPy.SimPlot import *plt = SimPlot()plt.plotBars([[0, 0], [1, 1], [2, 4], [3, 9]],

color="blue", width=2)plt.mainloop()

which results in:

And finally, a scatter diagram:

# Prog5.pyfrom SimPy.SimPlot import *plt = SimPlot()plt.plotScatter([[0, 0], [1, 1], [2, 4], [3, 9]],

color="green", size=2, marker='triangle')plt.mainloop()

and its output:



With a bit more involved programming, it is also possible to have several plots in one diagram and to have severaldiagrams in one Frame (just execute SimPlot.py to get these plots):



Note: In future versions of SimPlot, this part of the API will also be simplified so that it will require significantly lesscoding.

Plotting Monitor instances

Class Monitor is the prime data collection tool for SimPy simulations. SimPlot therefore caters for easy plotting fromMonitor intances.

Here is an example of a simple simulation using a Monitor:

# Monitorplot.pyfrom random import uniform



from SimPy.Simulation import *from SimPy.Recording import *from SimPy.SimPlot import *

class Source(Process):def __init__(self, monitor):

Process.__init__(self)self.moni = monitorself.arrived = 0

def arrivalGenerator(self):while True:

yield hold, self, uniform(0, 20)self.arrived += 1self.moni.observe(self.arrived)

initialize()moni = Monitor(name="Arrivals", ylab="nr arrived")s = Source(moni)activate(s, s.arrivalGenerator())simulate(until=100)

plt = SimPlot()plt.plotStep(moni, color='blue')plt.mainloop()

This produces:



5.1.3 Simple plotting API

Overview

A SimPlot application program has the following structure:

• instantiation of the SimPlot class

• call of one or more plotting methods

• call to the SimPlot instance’s mainloop method

SimPlot exposes plotting methods at two levels, a simple API with limited capabilities but ease of use, and an advancedAPI with SimPlot’s full capabilities, but with more involved, verbose programming.

This section deals with the simple API.

SimPlot

class SimPlot

This class provides the plotting capabilities.

plotLine

classmethod SimPlot.plotLine(values[, optional parameters])Generates a line plot from a list of tuples (or lists) or from a Monitor (any instance which has the attributes‘name’, ‘tlab’, ‘ylab’).

Parameters

• values (list or Monitor) – (mandatory) a list of two-element lists (or tuples), ora Monitor instance (any instance which has the attributes ‘name’, ‘tlab’, ‘ylab’)

• windowsize – the plotting window size in pixels (tuple); default: (500,300)

• title (string) – the plot title ; if values is a Monitor, Monitor name is used if no titlegiven

• width (integer or floating point) – line drawing width, default: 1

• color (Tkinter colour type) – line colour; default: ‘black’

• smooth (boolean) – if True, makes line smooth; default: “True”

• background (Tkinter colour type) – colour of plot background; default: ‘white’

• xlab (string) – ** : label on x-axis of plot; if values is a Monitor, Monitor tlab is taken;default: ‘x’

• ylab – label on y-axis of plot (string); if values is a Monitor, Monitor ylab is taken; default:‘y’

• xaxis – layout of x-axis (None = omit x-axis; ‘automatic’ = make x-axis at least longenough to include all x-values, using round values; ‘minimal’ = have x-axis go exactly fromminimal to maximal x-value provided; tuple (xlow,xhigh) = have x-axis go from xlow toxhigh); default: ‘automatic’

• yaxis – layout of y-axis (None = omit y-axis; ‘automatic’ = make y-axis at least longenough to include all y-values, using round values; ‘minimal’ = have y-axis go exactly from


https://docs.python.org/3/library/stdtypes.html#list

https://docs.python.org/3/library/string.html#module-string

https://docs.python.org/3/library/string.html#module-string


minimal to maximal y-value provided; tuple (ylow,yhigh) = have y-axis go from ylow toyhigh); default: ‘automatic’

Return type Reference to GraphBase object which contains the plot.

plotStep

classmethod SimPlot.plotStep(values[, optional parameters])Generates a step plot from a list of tuples (or lists) or from a Monitor (any instance which has the attributes ‘name’,‘tlab’, ‘ylab’). A horizontal line is drawn at a y-value until y changes, creating a step effect.

<variable> = <SimPlotInstance>.plotStep(values[,optional parameters])

param values (mandatory) a list of two-element lists (or tuples), or a Monitor instance (anyinstance which has the attributes ‘name’, ‘tlab’, ‘ylab’)

Optional parameters with defaults:

• windowsize=(500,300), : the plotting window size in pixels (tuple)

• title=”“ : the plot title (string); if values is a Monitor, Monitor name is used if no title given

• width=1 : line drawing width (integer or floating point)

• color=’black’ : line colour (Tkinter colour type; see section on Colours in SimPlot)

• background=’white’ : colour of plot background (Tkinter colour type; see section on Colours in SimPlot)

• xlab=’x’ : label on x-axis of plot (string); if values is a Monitor, Monitor tlab is taken

• ylab=’y’ : label on y-axis of plot (string); if values is a Monitor, Monitor ylab is taken

• xaxis=’automatic’ : layout of x-axis (None = omit x-axis; ‘automatic’ = make x-axis at least long enoughto include all x-values, using round values; ‘minimal’ = have x-axis go exactly from minimal to maximalx-value provided; tuple (xlow,xhigh) = have x-axis go from xlow to xhigh)

• yaxis=’automatic’ : layout of y-axis (None = omit y-axis; ‘automatic’ = make y-axis at least long enoughto include all y-values, using round values; ‘minimal’ = have y-axis go exactly from minimal to maximaly-value provided; tuple (ylow,yhigh) = have y-axis go from ylow to yhigh)

Return value: Reference to GraphBase object which contains the plot.

plotBars

Generates a bar chart plot from a list of tuples (or lists) or from a Monitor.

Call: <SimPlotInstance>.plotBars(values[,optional parameters])

<variable> = <SimPlotInstance>.plotBars(values[,optional parameters])

Mandatory parameters:

• values : a list of two-element lists (or tuples), or a Monitor instance



• title=’‘ : the plot title (string); if values is a Monitor, Monitor name is used if no title given

• width=1 : outline drawing width (integer or floating point)

• color=’black’ : outline colour (Tkinter colour type; see section on Colours in SimPlot)



• fillcolor=’black’ : colour with which bars are filled (Tkinter colour type; see section on Colours in Sim-Plot)

• fillstyle=’‘ : density of fill (default=100%; Tkinter bitmap)

• outline=’black” : colour of bar outline ((Tkinter colour type; see section on Colours in SimPlot)







plotHistogram

Generates a histogram plot from a Histogram or a Histogram-like list or tuple. A SimPy Histogram instance is a listwith items of two elements. It has n+2 bins of equal width, sorted by the first element, containing integer values ==the counts of the bins. The first bin is the ‘under’ bin, the last the ‘over’ bin. Histogram objects are produced fromMonitor objects by calling the Monitor method histogram().

Call: <SimPlotInstance>.plotHistogram(values[,optional parameters])

<variable> = <SimPlotInstance>.plotHistogram(values[,optional parameters])




• windowsize=(500,300) : the plotting window size in pixels (tuple)





• xlab=’x’ : label on x-axis of plot (string)

• ylab=’y’ : label on y-axis of plot (string)





plotScatter

Generates a scatter diagram plot from a list of tuples (or lists) or from a Monitor.

Call: <SimPlotInstance>.plotScatter(values[,optional parameters])

variable = <SimPlotInstance>. plotScatter(values[,optional parameters])






• marker=’circle’ : symbol type (literal; values supported: ‘circle, ‘dot’, ‘square’, ‘triangle, ‘trian-gle_down’, ‘cross’, ‘plus’)





• outline=’black’ : colour of marker outline ((Tkinter colour type; see section on Colours in SimPlot)







postscr

Saves Postscript output from a plot to a file. After e.g. aPlot=plotLine([0,1],[3,4]), aPlot.postscr("c:\\myplot.ps") outputs the line plot in Postscript to file c:\myplot.ps.

Call: <plotinstance>.postscr([optional parameter]) (with <plotinstance> being a reference to the GraphBase objectwhich contains the plot)



• “<filename>” : name of file (complete path) to which Postscript output is written. If omitted, a dialogasking the user for a filename pops up.

Return value: None.



5.1.4 Advanced plotting API

Overview

The advanced SimPlot API is more verbose than the simple one, but it offers more flexibility and power. The detailedstructure of a program using that API is:

1. make an instance of SimPlot (this initializes Tk and generates a Tk Toplevel container <instance>.root whichpoints at the Tk object.)

2. (optional) make other Tk container(s)

3. (optional) give the container a title

4. make one or more plot objects (the lines or other figures to plot)

5. put the plot objects into a GraphObject (this does the necessary scaling)

6. make a Tk container (e.g. a Frame widget) in the previous container (from step 1 or 2)

7. make a background (with title, axes, frame, etc.) in that container for the GraphObject to be drawn against (i.e.,create the graph widget and associate the GraphObject with it)

8. instruct the Tk geometry manager (pack or grid) where to put the background in the Tk container

9. draw the GraphObject against the background

10. instruct the Tk geometry manager concerning the container from step 6

11. (optional) save plot as Postscript file

12. start the Tk mainloop

An example:

# AdvancedAPI.pyfrom SimPy. SimPlot import *plt = SimPlot() # step 1plt.root.title("Advanced API example") # step 3line = plt.makeLine([[0, 42], [1, 1], [4, 16]]) # step 4bar = plt.makeBars([[0, 42], [1, 1], [4, 16]],

color='blue') # step 4sym = plt.makeSymbols([[1, 1]], marker="triangle",

size=3, fillcolor="red") # step 4obj = plt.makeGraphObjects([line, bar, sym]) # step 5frame = Frame(plt.root) # step 6graph = plt.makeGraphBase(frame, 500, 300,

title="Line and bars") # step 7graph.pack() # step 8graph.draw(obj) # step 9frame.pack() # step 10graph.postscr() # step 11plt.mainloop() # step 12

Which generates:



Clearly, this level API is more verbose, but allows putting several diagrams with different plot types into one plot, orputting putting several plots into one frame (side by side, vertically, or in table fashion).

title

Assign a title to appear in the container’s title bar. (This is a method exposed by a Tk Toplevel container.)

Call: <rootInstance>.title(title)


• title : the title to appear in the container’s title bar (string)

Optional parameters: None.

Return value: None.

makeLine

Generates a line plot object from a list of tuples (or lists).

Call: <variable> = <SimPlotInstance>.makeLine(values[,optional parameters])


• values : a list of two-element lists (or tuples)


• color = ‘black’ : line colour (Tk colour value)

• width = 1 : line width (integer or float)

• smooth = False : smooth line if True (boolean)

• splinesteps = 12 : number of spline steps for smoothing line (integer); the higher, the better the line followsthe points provided

Return value: Reference to a line plot object (GraphLine)



makeStep

Generates a line plot object from a list of tuples (or lists). A horizontal line is generated at a y-value until y changes,creating a step effect.

Call: <variable> = <SimPlotInstance>.makeStep(values[,optional parameters])




• color = ‘black’ : line colour (Tk colour value)

• width = 1 : line width (integer or float)

Return value: Reference to a line plot object (GraphLine)

makeSymbols

Generates a scatter diagram plot object with markers from a list of tuples (or lists).

Call: <variable> = <SimPlotInstance>.makeSymbols(values[,optional parameters])




• marker=’circle’ : symbol type (literal; values supported: ‘circle, ‘dot’, ‘square’, ‘triangle, ‘trian-gle_down’, ‘cross’, ‘plus’)





• outline=’black’ : colour of marker outline ((Tkinter colour type; see section on Colours in SimPlot)

Return value: Reference to a scatter plot object (GraphSymbols)

makeBars

Generates a bar chart plot object with markers from a list of tuples (or lists).

Call: <variable> = <SimPlotInstance>.makeBars(values[,optional parameters])




• width=1 : width of bars (integer or floating point)

• color=’black’ : bar colour (Tkinter colour type; see section on Colours in SimPlot)





• outline=’black” : colour of bar outline ((Tkinter colour type; see section on Colours in SimPlot)

Return value: Reference to a bar chart plot object (GraphSymbols)

makeHistogram

Generates a histogram plot from a Histogram or a Histogram-like list or tuple. A SimPy Histogram instance is a listwith items of two elements. It has n+2 bins of equal width, sorted by the first element, containing integer values ==the counts of the bins. The first bin is the ‘under’ bin, the last the ‘over’ bin. Histogram objects are produced fromMonitor objects by calling the Monitor method histogram().

Call: <variable> = <SimPlotInstance>.makeBars(values[,optional parameters])


• values : a Histogram-like object


• width=1 : width of line (integer or floating point)


makeGraphObjects

Combines one or mor plot objects into one plottable GraphObject.

Call: <variable> = <SimPlotInstance>.makeGraphObjects(list_of_plotObjects)


• list_of_plotObjects : a list of plot objects


Return value: Reference to a GraphObject

makeGraphBase

Generates a canvas widget in its Tk container widget (such as a Frame) with the plot’s background (title, axes, axislabels).

Call: <variable> = <SimPlotInstance>.makeGraphBase(master, width, height [,optional parameters])


• master : container widget for graph widget

• width : width of graph widget in pixels (positive integer)

• height : height of graph widget in pixels (positive integer)


• background=’white’ : colour of plot background (Tk colour value)

• title=”“ : title of plot (string)

• xtitle=’‘ : label on x-axis (string)

• ytitle=’‘ : label on y-axis (string)



Return value: Reference to a GraphBase object (graph widget)

pack

Controls how graph widget is arranged in its master container. (Inherited from Tk Packer geometry manager.)

Call: <GraphBaseInstance>.pack([optional parameters])



• side : where to place graph widget (side=LEFT: to the left; side=TOP: at the top; Tk Packer literals)

• fill : controls whether graph fills available space in window (fill=BOTH: fills in both directions; fill=X:horizontal stretching; fill=Y: vertical stretching)

• expand=NO : controls whether Packer expands graph widget when window is resized (expand=TRUE:widget may expand to fill available space)

Return value: None

draw

Draws the plot background and the lines/curves in it.

Call:

<GraphBaseInstance>.draw(graph,[optional parameters])


• graphics : graph widget (GraphBase) instance


• xaxis=’automatic’ : controls appearance of x-axis (None: no x-axis; “minimal”: axis runs exactlyfrom minimal to maximal x-value; “automatic” : x-axis starts at 0 and includes maximal x-value; tuple(xlow,xhigh) = have x-axis go from xlow to xhigh)

• yaxis=’automatic’ : controls appearance of y-axis (None: no y-axis;”minimal”: axis runs exactly fromminimal to maximal y-value; “automatic” : y-axis starts at 0 and includes maximal y-value; tuple(ylow,yhigh) = have y-axis go from ylow to yhigh

Return value: None

postscr

After call to draw , saves Postscript output from a plot to a file.

Call: <GraphBaseInstance>.postscr([optional parameter])



• “filename” : name of file (complete path) to which Postscript output is written. If omitted, a dialog askingthe user for a filename pops up.

Return value: None.



5.1.5 Colours in SimPlot

Colours in SimPlot are defined by Tk. The Tk colour model is RGB. The simplest way to identify a colour is to useone of the hundreds of Tk-defined literals such as “black”, “aquamarine”, or even “BlanchedAlmond”. See the Tkcolour page for definitions.

5.2 SimGUI Manual

Authors Klaus Muller <[email protected]>

SimPy Classic Release 2.3.3



Date December 2011


Contents

• SimGUI Manual

– Acknowledgements

– Introduction

– SimGUI overview

– The SimGUI API

– Miscellaneous SimGUI capabilities

This manual describes SimGUI, a GUI framework for SimPy simulation applications.


The initial ideas for using a Tk/Tkinter-based GUI for SimPy simulation models and first applications came from MikeMellor and Prof. Simon Frost of University of California, San Diego. Simon has been a very productive co-developerof SimGUI.

Following an idea by Simon Frost, SimGUI uses a great Tkinter-based console for conversing with the Python inter-preter, developed by Ka-Ping Yee <[email protected]>.

5.2.2 Introduction

SimGUI is a GUI framework for SimPy simulation programs. It provides for:

• a standard layout of the user interface, including its menu structure

• running a simulation

• viewing simulation output

• saving the output to a file

5.2. SimGUI Manual 351

../../_static/Tkcolors.htm

../../_static/Tkcolors.htm





• changing the simulation model parameters

• viewing a model description

• viewing the simulation program’s code

• a Python console for debugging

SimGUI is based on the Tk GUI library.

5.2.3 SimGUI overview

Here is a minimalistic program which does nothing but show the SimGUI user interface:

## SimGUIminimal.pyfrom SimPy.SimGUI import *root=Tk()gu=SimGUI(root,consoleHeight=20)gu.mainloop()

Running it produces this output:

The large frame is the SimGUI application window. To its right are the standard menu items of SimGUI. To showthem, the menus have been torn off by clicking on the dotted line on all SimGUI drop-down menu items.

The SimGUI application window consists of five widgets:

• the outside frame is a Tk Toplevel widget with a default title (which can be changed),

– a menu bar, a Tk Menu widget which can be adapted by the application program (contained in toplevel)

– the output window, a Tk Frame widet for SimPy program output (contained in toplevel)

* the status bar for one-line status messages, a Tk Label widget (contained in output window)

* the output console for application output, a Tk Text widget (contained in output window)



The File sub-menu is for saving or opening files and also for quitting the application. By default, it supports savingthe content of the output console.

The Edit sub-menu is for any editing or change operations. By default, it supports changing model parameters andclearing the output console.

The Run sub-menu is for running the SimPy model. By default, it is empty, as there is no model to run.

The View sub-menu is for calling up any application output, e.g. plots or statistics. By default, it has the ability toautomatically show the data from any Monitor instance in the SimPy application.

The Help sub-menu is for viewing any information about SimGUI and the specific application behind it. By default,it provides information about the version of SimGUI, lists the application’s code and allows the launching of a Pythonconsole for testing and debugging purposes.

To show how simple it is to interface a model to SimGUI, here is an example with a simple simulation model:

#!/usr/bin/env python__doc__=""" GUIdemo.pyThis is a very basic model, demonstrating the easeof interfacing to SimGUI."""from __future__ import generatorsfrom SimPy.Simulation import *from random import *from SimPy.SimGUI import *

class Launcher(Process):nrLaunched=0def launch(self):

while True:gui.writeConsole("Launch at %.1f"%now())Launcher.nrLaunched+=1gui.launchmonitor.observe(Launcher.nrLaunched)yield hold,self,uniform(1,gui.params.maxFlightTime)gui.writeConsole("Boom!!! Aaaah!! at %.1f"%now())

def model():gui.launchmonitor=Monitor(name="Rocket counter",

ylab="nr launched",tlab="time")initialize()Launcher.nrLaunched=0for i in range(gui.params.nrLaunchers):

lau=Launcher()activate(lau,lau.launch())

simulate(until=gui.params.duration)gui.noRunYet=Falsegui.writeStatusLine("%s rockets launched in %.1f minutes"%

(Launcher.nrLaunched,now()))

class MyGUI(SimGUI):def __init__(self,win,**par):

SimGUI.__init__(self,win,**par)self.run.add_command(label="Start fireworks",

command=model,underline=0)self.params=Parameters(duration=2000,maxFlightTime=11.7,nrLaunchers=3)

root=Tk()gui=MyGUI(root,title="RocketGUI",doc=__doc__,consoleHeight=40)gui.mainloop()



It produces the following output when the model run command is selected:

class MyGUI adds one menu item under the Run menu. It also defines three parameters (duration, maxFlighTime andnrLaunchers) with their initial values which the user can change interactively before each run. The MyGUI instancenamed gui sets the window title, the model description (the __doc__ string) and the height of the output window.

The simulation part of the program writes to the output console and to the status line.

The model uses a Monitor for keeping track of the number of rockets launched over time. Because it is made anattribute of the MyGUI instance, the super class (SimGUI) can output the Monitor after a run. This requires noapplication code. When the menu item ‘Collected data’ under the View menu is selected, this results in:



5.2.4 The SimGUI API

Structure

The SimGUI module exposes the following API to the applications programmer:

class SimGUI(object)def __init__

self.doc = docself.noRunYet=Trueself.top = Menuself.file = Menuself.edit = Menuself.run = Menuself.view = Menu



self.help = Menudef makeFileMenudef makeEditMenudef makeRunMenudef makeViewMenudef makeHelpMenudef writeConsoledef saveConsoledef clearConsoledef writeStatusLinedef aboutdef notdonedef showTextBoxdef mainloop

class Parametersdef __init__def show

class SimGUI, __init__ constructor

Encapsulates the SimGUI functionality.

Call: <variable>.SimGUI(win[,optional parameters])


• win : the master widget in which the SimGUI widgets are embedded


• title=SimGUI : the title of the top level window (string)

• doc=None : the model description (string)

• consoleHeight=50 : the height of the output console in lines (positive integer)

Return value: Returns a reference to the SimGUI instance.

SimGUI instance fields

In addition to the constructor parameters, the SimGUI fields of interest to the applications programmer are:

• self.noRunYet=True : a predicate indicating whether the model has been run yet; must be set to True after eachmodel run; should be tested by application program before any run-dependent output is produced (boolean)

• self.top = Menu : the top level menu widget (menu bar)

• self.file = Menu : the ‘File’ menu widger

• self.edit = Menu : the ‘Edit’ menu widget

• self.run = Menu : the ‘Run’ menu widget

• self.view = Menu : the ‘View’ menu widget

• self.help = Menu : the ‘Help’ menu widget



Adding menu items

Menu items can be added to SimGUI submenus by:

<menufield>.add_command(label= <string,command=<callable>, underline=<integer>)

E.g. self.run.add_command(label=”Start fireworks”,command=model,underline=0). This is all standard Tk/Tkintersyntax – read any Tk/Tkinter manual or book.

Changing menus

Any of the submenus provided by SimGUI can be replaced by overloading one or more of the methods makeFile-Menu, makeEditMenu, makeRunMenu, makeViewMenu, makeHelpMenu. This is done by defining the methodsto be overloaded in the SimGUI subclass.

The overloading method should look like:

def makeFileMenu():self.file = Menu(self.top)self.file.add_command(label='Save some results',

command=self.saveResults, underline=0)self.file.add_command(label='Get out of here',

command=self.win.quit,underline=0)self.top.add_cascade(label='File',menu=self.file,underline=0)

Note: It is advisable to keep the basic the SimGUI menu structure in order to maintain the SimGUI look-and-feel.

writeConsole

Writes a string into the output console Text widget, with newline at end.

Call:

<SimGUI instance>.writeConsole(text)



• text=’‘ : text to write (string)

Return value: None.

saveConsole

Saves output console to a text file which the user is prompted to name.

Call: <SimGUI instance>.saveConsole()



Return value: None



clearConsole

Clears output console.

Call: <SimGUI instance>.clearConsole()



Return value: None

writeStatusLine

Writes a one-line string to the status line.

Call:

<SimGUI instance>.writeStatusLine(text)



• text=’‘ : text to write (string, not longer than 80 character)

Return value: None.

notdone

Brings up a warning box staing that a feature is not implemented yet. Useful during development of application.

Call:

<SimGUI instance>.notdone()



Return value: None.

showTextBox

Pops up a text box (Text widget)with a text in it.

Call:

<SimGUI instance>.showTextBox(optional parameters)



• title=’‘ : title of text box (string)

• text=’‘ : text to write (string)

• width=80 : width of box in characters (positive integer)

• height=10 : height of box in lines (positive integer)

Return value: None.



mainloop

Starts up SimGUI execution.

Call: <SimGUI instance>.mainloop()



Return value: None

class Parameters, __init__ constructor and adding parameters

This class provides for interactive user changes of model parameters. Any user-input is checked against the type ofthe original (default) value of the parameter.In this version, parameters of type integer, floating point, text and list aresupported. Boolean parameters can be implemented by using 0 for False and 1 for True.

Example: gui.params=Parameters(endtime=2000, numberCustomers=50, interval=10.0, trace=0)

This results in parameters gui.params.numberCustomers, gui.params.interval and gui.params.trace.

Call: <SimGUI instance>.params=Parameters(par) (constructor) <SimGUI in-stance>.params.<name>=<value> (adding parameters)


• par : one or more pairs <name>=<value>, separated by commas. <value> must be one of the four typessupported.

• <name> : the parameter name (any legal Python identifier)

• <value> : the parameter’s initial value (must be one of the four types supported, i.e. integer/boolean,floating point, text and list)


Return value: A Parameter instance.

show

Returns the parameter name/value pairs of a Parameter instance in pretty-printed form (one pair per line).

Call: <Parameter instance>.show()



Return value: A string with the name/value pairs separated by newline (‘n’).

5.2.5 Miscellaneous SimGUI capabilities

Source code listing

The ‘Model code’ item of the ‘Help’ submenu lists the application code of a running SimGUI application by outputtingthe content of sys.argv[0]. No user programming is required for this.



Automatic display of Monitor instances

After a model run, any Monitor instance which is referred to by a SimGUI (sub)class instance is shown in the outputconsole by the ‘Collected data’ item on the ‘View’ submenu. Just e.g. gui.waitMon=Monitor(“Waiting times”),where gui refers to a SimGUI (sub)class instance, is enough to facilitate this.

$Revision: 297 $ $Date: 2009-03-31 02:24:46 +1300 (Tue, 31 Mar 2009) $ kgm

5.3 Visualize model events and states using SimulationGUIDebug

Brian Jacobs, Kip Nicol and Logan Rockmore, a group of senior students of Professor Matloff at U. of California atDavis have developed a great tool for gaining insight into the event-by-event execution of a SimPy model. It is equallyuseful for teaching and program debugging.

The SimulationGUIDebug tool is a SimPy module which shows a GUI with a model’s event list and the statusof selected Process and Resource instances over time. The user can interactively step from event to event and setbreakpoints at one or more points in (simulated) time.

SimulationGUIDEbug can be used as a standalone debuger or together with Python debuggers such as PDB.

The tool was originally written for SimPy 1.9. It has been ported to SimPy 2.0.

The tool’s documentation (user guide, design overview, etc.) is here .


CHAPTER 6

Acknowledgments

SimPy 2.3.3 has been primarily developed by Stefan Scherfke and Ontje Lünsdorf, starting from SimPy 1.9. Theirwork has resulted in a most elegant combination of an object oriented API with the existing API, maintaining full back-ward compatibility. It has been quite easy to integrate their product into the existing SimPy code and documentationenvironment.

Thanks, guys, for this great job! SimPy 2.0 is dedicated to you!

The many contributions of the SimPy user and developer communities are of course also gratefully acknowledged.

361


362 Chapter 6. Acknowledgments

CHAPTER 7

Indices and tables

• genindex

• search

363


364 Chapter 7. Indices and tables

Index

Aabandoning, 257

Level, 45abandoning (reneging), 255, 304activate, 19Advanced synchronization/scheduling commands, 263,

297allMonitors

monitor, 232asynchronous

interrupt, 22

Bbalking, 255, 304bank01_OO, 237, 279bank02, 239bank02_OO, 282bank03, 240bank03_OO, 284bank05, 238bank05_OO, 281bank06, 242bank06_OO, 285bank07, 244bank07_OO, 287bank08, 245bank08_OO, 289bank09, 247bank09_OO, 290bank10, 248bank10_OO, 292bank11, 258bank11_OO, 294bank20, 252bank20_OO, 301bank21, 257bank23, 253bank23_OO, 302bank24, 255

bank24_OO, 304

Ccancel, 21car wash

example Store, 51cause

interrupted, 23count

monitors, 231current state

monitors, 232

Eentity

creation, 17Error Messages, 59establish

Histogram Monitor, 58Histogram Tally, 57

examplebreakdown, 24, 97bus, 24car, 92, 226cars, 94, 228carsT, 94carT, 93carwash, 96diffpriority, 35, 98firework, 21, 99Histogram Monitor, 58Histogram Tally, 57Level, 43levelinventory, 100master/slave, Store, 51message, 18, 102monitor, 58, 102parcel, Store, 51resource, 103, 228resource with tracing, 233

365


resourcemonitor, 39, 104Romulans, 29, 106shopping, 19, 107showing waiting, 27source, 22Store car wash, 51Store yield get with filter, 47storewidget, 108tally, 57, 109, 110tally2, 57wait_or_queue, 27

Ffired processes

list, 26function

model, 261, 296

GGathering statistics, 258, 294get in order

yield, 50Glossary, 61Graphical Output, 263, 297GUI input, 263, 297

HHistogram

Monitor establish, 58Monitor, example, 58Tally establish, 57Tally, example, 57

histogrammonitors, 232

Histograms, 56hold

yield, 19, 225

Iinterrupt, 23interruptCause, 23interrupted, 23

cause, 23interruptedCause, 23interruptLeft, 23

LLevel, 42

abandoning, 45definition, 42get, 43reneging, 45

Levels, 235, 279

definition, 14levels

definition, 221

MM/M/1

queue, 245, 289master/slave

Store example, 51mean

monitors, 231Tally, 55

modelfunction, 261, 296

Monitor, 15, 53define, 54establish, Histogram, 58example Histogram, 58mean, 55observe, 54special methods, 55statistics, 55timeAverage, 55

monitorallMonitors, 232startCollection, 232

Monitoredqueue, 258, 294

monitoringresource queues, 232

Monitors, 230, 235, 258, 279, 294monitors

count, 231current state, 232data summaries, 231defining, 230histogram, 232mean, 231observe, 231reset, 231timeVarience, 232total, 231tseries, 231var, 231yseries, 231

Multiple runs, replications, bank12, 261Multiple runs, replications, bank12_OO, 296

Oobject creation

process, 17Object Oriented interface, 16object-oriented style, 237observe

366 Index


Tally, 54observing data, 231Other forms of Resource Facilities, 263, 297

Ppair monitors

timeAverage, 231parcel

Store example, 51passivate

yield, 21, 226PEM

Process Execution Method, 236, 238, 281plotLine() (SimPlot class method), 342plotStep() (SimPlot class method), 343preemptable

Resource, 34Priorities and Reneging, 263, 297priority, 251, 300

Resource request, 34priority: preemption, 253, 302PriorityQ

Resource, 34Process, 16

defining, 17process

creation, 224defining, 223object creation, 17sleep, 226start, 225starting, 224

Process Execution MethodPEM, 17, 236, 238, 281

process objectactivation, 19, 224start, 20

Processes, 235, 279processes, 223

QqType

Resource, 34queue

M/M/1, 245, 289Monitored, 258, 294Resource, 244, 286

queue orderResource, 32

queued and fired processes, 27queued processes

find, 26queueevent

yield, 26

queueingSimevents, 25

Rrandom arrival, 238, 252, 281, 301random number generation, 229Random Number Seed, 261, 296Random numbers, 53Random service time, 245, 289reactivate, 21recording average waiting times, 259Recording Simulation Results, 230release

yield, 32, 228reneging, 255, 257, 304request

priority, Resource, 34yield, 32, 228

resetmonitors, 231

Resource, 245, 247, 289, 290defining object, 31monitor queue lengths, 39non-priority queueing, 32preemptable, 34preemptive request pattern, 35priority, 34priority queueing, 34PriorityQ, 34qType, 34queue, 244, 286queue order, 32queues, 32release, 32reneging, 37request, 32request priority, 34Several queues, 248, 292waitQ, 32

resourceactiveQ, 228actMon, 228attributes, 227capacity, 227define, 227example, 228monitored, 227n, 227name, 227unitName, 227waitMon, 228waitQ, 228

resource monitoring, bank15, 260resource queue

Index 367


monitoring, 232Resources, 14, 30, 235, 279resources, 221, 227

SService counter, 244, 287several counters, 247, 290Several queues

Resource, 248, 292Signalling, 25signalling

SimEvent, 25SimEvent

creating, 25signalling, 25

SimEventswaiting, 25

Simeventsqueueing, 25

SimPlot (built-in class), 342SimPy Process States, 60Simulation, 279SimulationTrace, 233sleep, 21sleep a process, 226source

example, 22Source of entities, 240, 284start

process, 225startCollection

monitor, 232Statistical Output, 263, 297statistics, 258, 294

Tally, 55Store, 45

car wash, example, 51definition, 46example master/slave, 51example parcel, 51yield get with filter, example, 47

Stores, 235, 279definition, 14

storesdefinition, 222

TTally, 53

define, 54establish, Histogram, 57example Histogram, 57mean, 55observe, 54printing a histogram, 57

statistics, 55timeAverage, 55

Tallydefinition, 15

Tallys, 235, 279timeAverage

Tally, 55timeVarience

monitors, 232total

monitors, 231tseries

monitors, 231

Vvar

monitors, 231

Wwait

waituntil, 28waitenvent

yield, 26waiting processes

find, 26waitQ

Resource, 32waituntil

yield, 29

Yyield

cancel, 21get, 47get in order, 50get with filter, 47hold, 19, 225passivate, 21, 226put, 47put in order, 50put with reneging, 49queueevent, 26reactivate, 21release, 32, 228release a resource, 228request, 32, 228request a resource, 228request with reneging, 37request with reneging after a time limit, 37request with reneging after an event, 38waitenvent, 26waituntil, 29

yielddefinition, 17

368 Index


yield get with filterexample Store, 47

yseriesmonitors, 231

Index 369

Documents

SimPy Documentation